CO m ) — co ~ co \ tz CO NOlinillSNI NVINOSH1IIAIS S3IUVdai1 LIBRARIES SMITHSONIAN INSTITUTION NOIlfUllS Z co 2: ^ . CO 2; .... CO to (ffc' ^1) E/5 CO CO CO Eo rvx- 2E ^ > x^uasTix * > CO z CO * 2 CO v z LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHimS S 3 I B V B 9 l~l LIBRARII co X co — CO NounuiSNi NvmosHims saiBvaan libraries Smithsonian institution Noiinius 1“ v Z r~ Z r- z m v V^v* -5- xjvasv^x m x^osvPz m co ' ± co £ co ™ LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IWS S3IBVBail LIBRARII co z "v CO Z CO z s ^ < /^*qX 2 < __ . z —\ W&sfjfa- 2 /SffwA H \ co s|t ^§j) co \N co co *351 co t « § A‘/; t Xw S ' V^v > ' 2 >' 2E Jo z co — x co NOlinillSNI _ NVIN0SH1IWS S3 I BVB8 II LIBRARIES SMITHSONIAN INSTITUTION NOlinillS co ~ co zz co O X^yosvA^X ~ O X^vos^x X -J Z _j Z -J LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NV1N0SH1MS S3ldVd9ll LIBRARII z m w '^omsjix m ^ m NOlinillSNI NVIN0SH1IIAIS S3lBVBan~LIBRARI E S ^SMITHSONIAN “ INSTITUTION ^ NOlinillS z co z co 2: ..... co ■H X /Pi CO O > w 2 v > x^r 7i&y ^ > CO x CO * z — Jo V z LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHUWS S3IBVB8I1 LIBRARII co zz co — CO NOlinillSNI NVINOSHimS S3 I BVBa 11 LIBRARIES SMITHSONIAN INSTITUTION NOlinillS “ r- z r- vSt *£&\ W x 2 2 CO 2 £ 73 > XJ — M'fc-WM*’' vVT m x^yost^ co co Zz co LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IWS S3IBVB9I1 L! B R A R I E ^ ^ ^ x co z ^ z ^ /rdvrrr(7>v X'O-tIJ-SgX ^ 2 z co A z co v z co ^ THSONIAN INSTITUTION NOIlDlllSNI NVINOSH1IWS S3IUVdan LIBRARIES SMITHSONIAN_ co ~ CO ~ (S) —p UJ CO Q __ X^jiixS^Z O v _ o \i^N_DVZ _ O ^osnims^sa 1 dvd a nJu b rar i es^ Smithsonian-" institution ^NoiiniusNi^NviNOSHims z r~ z n z ~ O zZatTtTT^v Xxa O _ Zoa O y CD X) > X) 6/ — m x^osvavz zi m \^vosv^>' " m = co — _ co _co THSONIAN INSTITUTION NOIXDXIXSNI NVIN0SH1I1NS S3iyvyaM LIBRARIES SMITHSONIAN z . to z to z z m. .vx —a z /p^nr^oX -h wmA w z vx -1 z ±7 o _ i/ ? 5 ' >' 2 X^osv^z >’ ^ ^ NosHims^SB 1 avaa n2u brar 1 es^smithsonian ^institution NoiinniSNi__NviNOSHims CO — CO — CO -7> ITHSONIAN INSTITUTION NOliOXIXSWI NVINOSHXIWS S3iyvyaiT LIBRARIES SMITHSONIAN Z> £ ^ m n^du/ N^oin^z m '*& W" 5 v > Xionxs^z ^ Z CO z 00 v Z co THSONIAN INSTITUTION NOIXDXIXSNI NVINOSHXIWS S3iyVBan LIBRARIES SMITHSONIAN co — co z co O _ N«uuj2X o — Ngimsgx o xyv pc^y — O N0SHxmszS3 1 yvy a n_lu brar i esz Smithsonian^ institution ZNoixnxixsNi7!NvmosHiuMS^ ji ® 1 J09 * I s ^ vx i= t; * /f^M\ t fSgjfc. g ^ It jl > |e CO m ^V^/Aj^c W'£' CO tVAS m x^x CO m CO Zl CO ITHSONIAN INSTITUTION NOIlfUllSNI NVINOSHJLIWS S3IUVHan LIBRARIES SMITHSONIAN co z co z > CO z < V s ^ Xo i-.j J The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 16 SPRING 1988 NUMBER 1 THE JOURNAL OF ARACHNOLOGY EDITOR: James E. Carico, Lynchburg College ASSOCIATE EDITOR: Jerome S. Rovner, Ohio University EDITORIAL BOARD: J. A. Coddington, National Museum of Natural History, Smithsonian Institution; J. C. Cokendolpher, Texas Tech University; F. A. Coyle, Western Carolina University; C. D. Dondale, Agriculture Canada; W. G. Eberhard, Universidad de Costa Rica; M. E. Galiano, Museo Argentino de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia, Missouri; N. F. Hadley, Arizona State University; N. V. Horner, Midwestern State University; H. W. Levi, Harvard University; E. A. Maury, Museo Argentino de Ciencias Naturales; M. H. Muma, Western New Mexico University; N. I. Platnick, American Museum of Natural History; G. A. Polis, Vanderbilt University; S. E. Riechert, University of Tennessee; A. L. Rypstra, Miami University, Ohio; M. H. Robinson, U.S. National Zoological Park; W. A. Shear, Hampden-Sydney College; G. E. Stratton, Albion College; W. J. Tietjen, Lindenwood College; G. W. Uetz, University of Cincinnati; C. E. Valerio, Universidad de Costa Rica. THE JOURNAL OF ARACHNOLOGY (ISSN 0161-8202) is published in Spring, Summer, and Fall by The American Arachnological Society at Texas Tech Press. Individual subscriptions, which include membership in the Society, are $25.00 for regular members, $15.00 for student members. Institutional subscriptions to The Journal are $40.00. Correspondence concerning subscriptions and member- ships should be addressed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. Susan E. Riechert, Department of Zoology, University of Tennessee, Knoxville, TN 37916 U.S. A., at $10.00 for each number; the Index to Volumes 1-10 is available for $10.00. Remittances should be made payable to The American Arachnological Society. Correspon- dence concerning undelivered issues should be addressed to the Texas Tech Press, Texas Tech University, Lubbock, Texas 79409, U.S. A. Change of address notices must be sent to the Membership Secretary. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without benefit of review. Manuscripts and all related correspondence must be sent to Dr. Jerome S. Rovner, Associate Editor, Department of Zoological Sciences, Ohio University, Irvine Hall, Athens, Ohio 45701, U.S. A. PROOFS, REPRINTS, and CHARGES: Authors will receive a reprint order form along with their proofs. Reprints are billed at the printer’s current schedule of costs. All authors will be billed by the printer for page charges. The charge per journal page will be assessed as follows: $20.00-nonmembers; $ 15.00-members acknowledging grant, institutional, or other support; $ 10.00-full members who do not have (and do not acknowledge) support; $5. 00-student members who do not coauthor with full members and who do not acknowledge support. Members may have page charges reduced for a limited number of pages; a statement of justification for reduction should accompany the manuscript when it is first sent to the Associate Editor for review. Bruce, J. A. and J. E. Carico. 1988. Silk use during mating in Pisaurina mira (Walckenaer) (Araneae, Pisauridae). J. Arachnol., 16:1-4. SILK USE DURING MATING IN PISAURINA MIRA (WALCKENAER) (ARANEAE, PISAURIDAE) ABSTRACT John A, Bruce and Janies E. Carico1 Biology Department Lynchburg College Lynchburg, VA 24501 USA The mating behavior of the nursery web spider Pisaurina mira is described for the first time. These spiders mate while suspended from draglines as in Oxyopes heterophthalmus (Latr.) and Peucetia viridans (Hentz). The unique feature of the mating in P. mira, however, is the male’s use of a veil of silk to wrap the female’s legs I and II into a flexed position prior to copulation while her legs III and IV are held in a flexed position by the male’s embrace. Mating is accomplished in a version of position II with body axes in a right angle as the female is cradled by the male’s legs. The use of silk to “tie” the female is reported elsewhere only in Xysticus. INTRODUCTION Pisaurina mira (Walckenaer) is known to use silk in the construction of the nursery, egg sac, and juvenile web (Carico 1972, 1985). The purpose of this paper is to describe for the first time the mating sequence of P. mira which includes the unique use of silk to bind the female’s legs I and II during copulation. MATERIALS AND METHODS Observations were made in the laboratory on spiders that were collected at night in Lynchburg, Virginia, and in Fairhaven, New Jersey, during the months of May and June during 1984-85. One field observation of mating revealed no differences from those observed in the laboratory. Male and immature female spiders were kept in glass jars and plastic containers begween mating bouts. Mature females were kept in separate aquaria, and some matings were observed in these containers. All spiders were fed blowflies ( Sarcophaga sp.) and houseflies ( Musca domestica) on alternate days. To observe and record the mating sequence in the laboratory, a non-enclosed mating arena was devised using potted philodendron ( Philodendron scadens oxycardium) or English ivy {He dr a helix) set in large, shallow trays of water. The water discouraged escape, and thus made possible an unobstructed view of behavior. Events were recorded with color video cameras and 35 mm cameras. ‘Send correspondence to JEC. 2 THE JOURNAL OF ARACHNOLOGY RESULTS Communication before mating. — In a typical observational session, the female was first introduced onto the plant, upon which she subsequently laid down a series of draglines, by moving upward and across the outer parts of the plant. The male was introduced onto the plant 15 minutes to 3 hours after her wandering activity subsided. He wandered randomly across the plant, climbing and descending until he contacted her dragline. He then followed the dragline, passing the prolateral surface of each palpus alternately over it in a lycosid-like manner, similar to that reported for Lycosa rabida Walckenaer (Tietjen and Rovner 1980). During the process of his trail-following behavior, the male periodically stopped, released the dragline, and raised and extended either leg I. The time duration of each pause (30 sec-5 min) increased inversely with the distance to the female (20.3 cm-2.5 cm). There was no visible response by the female. Gradually, the distance between members of the mating pair decreased until the male touched her hind legs with his legs I and II, resulting in leg interplay between the sexes. In each instance, the tarsal portion of the male’s leg made contact with her tibiae, metatarsi and sometimes patellae. Mating. — When she was not receptive, the female climbed a short distance and descended immediately on a line to the lower branches of the plant. When she was receptive, the pair remained in contact while in a stationary position for a period of 5-20 sec. She then climbed the remaining distance to the mating area (a leaf or stem) where she attached a dragline and moved to a position beneath. Without pause, he approached from her posterior and moved to a location directly above her. He moved laterally across the edge of his perch as she drew her legs I and II against her carapace and descended freely (1.6-3. 8 cm) on her dragline. She was thus suspended free in space, face downward, inactive (possibly in a cataleptic state) and tethered only by a dragline (Fig. 1). He descended after her on his dragline, while following her dragline with his leg I. He moved across the dorsal surface of her abdomen, using his legs I and II and palps to rotate her as he moved to her ventral surface. As he rotated her three to five times, he pulled a veil of silk across her legs and bound them in a flexed position (Fig. 2). After he completed the veil, he attached his dragline to her legs I and II, which left her suspended on two draglines. To prepare for copulation he cradled her body with his legs, “folded” her legs III and IV into a flexed position, and assumed a version of position II with body axes at right angles (Fig. 3). While in this position he paused to pass his palps through his chelicerae to moisten them before insertion. The left palpus was then applied to the left atrium of her epigynum, and, after shifting his body to the other side, the right palpus was applied to the right atrium. The palpal bulb remained expanded 20-30 sec during each insertion of the embolus. There was a total of three to five insertions with a shift of the body between each insertion. The female became increasingly active during the final insertion, which indicated that the state of receptivity had ended. He discontinued cradling her with his legs, and her legs III and IV assumed an extended position. He wrapped her legs I and II with an additional veil of silk, climbed over her ventral surface and onto her dragline, leaving her bound and suspended on the line. He retreated to the lower portion of the plant while she freed herself from the silken veil and descended on her line to another location on the plant. BRUCE AND CARICO — PISA URINA MIRA MATING 3 Figs. 1-3. — Mating in Pisaurina mira: 1, female (shaded) suspended on a dragline while the male approaches from above; 2, male rotates female and wraps her legs I and II with a veil of silk; 3, pair suspended in copula with male holding legs III and IV of female in a flexed position. DISCUSSION Because of the prevailing tendency towards agonistic behavior among spiders — even conspecifics — it seems likely that binding of the female’s legs with silk prior to copulation, an apparent advantage for the survival of the male, would be a widespread phenomenon among spiders. To our knowledge, however, wrapping of the female with silk prior to copulation has been described previously only for Xysticus (Bristowe 1958). (However, female wrapping by the male has been recently discovered in a Neotropical pisaurid, Ancylometes hogotensis ; Merrett 1988). In Xysticus , the male spins a “bridal veil” of silk across the legs and body of the female to bind her to the substratum, but in R mira , binding of the female occurs in a wrapping fashion while she is suspended free from a dragline. The known differences in detail of the behaviors, along with the well-known 4 THE JOURNAL OF ARACHNOLOGY morphological differences between these two spiders, suggest that silk binding of the female was probably derived separately. The aspect of mating while suspended in space on a dragline is known in some species of Oxyopidae. The descriptions of mating in Oxyopes heterophthalmus (Latreille) (Gerhardt 1933) and Peucetia viridans (Hentz) (Whitcomb and Eason 1965) agree with our observation of P mira (but with the absence of silk wrapping). In particular, P. mira shows the rotation, or “twirling” of the female that is described for R viridans. This agreement may help support the conclusion by Brady (1964) on morphological grounds that there may be a close phylogenetic relationship between the Oxyopidae and Pisauridae. The wrapping of the female’s legs causes, at most, a brief restraint of the female as she becomes active following mating. As a result of subsequent repeated attempts at mating by the male, the additional silk causes the female to be further immobilized. The functional outcome of this wrapping procedure may serve to reduce predation on the male. Because the female’s legs are free from any substrate during mating, contrary to the case in most mating spiders, it seems probable that the male’s body would be the first object that the female would contact, and therefore, would immediately place him in jeapardy. Having the female’s legs restrained, even for a short time, provides what may be the critical opportunity for the male to escape predation by his mate. Published work on mating behavior in other pisaurids is known to us only for the genera Dolornedes and Pisaura , and the features of suspension and wrapping of the female are not included in any of these. We suggest that the mating behavior in representative species of other pisaurid genera be investigated for this character along with other characters which could aid in a better understanding of the systematics of this complex family. ACKNOWLEDGMENTS We thank Carolynn Bruce who contributed the illustrations and Nelle Carico and Peter Merrett who reviewed the manuscript. REFERENCES CITED Carico, J. E. 1972. The Nearctic spider genus Pisaurina (Pisauridae). Psyche, 79(4) 295-3 10. Carico, J. E. 1985. The description and significance of the juvenile web of Pisaurina mira (Walck.) (Araneae: Pisauridae). Bull. British Arachnol. Soc., 6(7): 295 •■296, Brady, A. R. 1964. The lynx spiders of North America, north of Mexico (Araneae: Oxyopidae). Bull. Mus. Comp. Zooh, 131(13):429-518. Bristowe, W. S. 1958. The World of Spiders. Collins. London. Gerhardt, U. 1933. Neue Untersuchungen zur Sexualbiologie der Spinnen, insbesoedere an Arten der Mittelmeerlander und der Tropen. Z. Morphol. Okol. Tiers. , 27(1): 1-75. Merrett, P. 1988. Notes on the biology of the neotropical pisaurid, Ancylomeies bogotensis (Keyserling) (Araneae: Pisauridae). Bull. British Arachnol. Soc., 7(7):(in press). Tietjen, W. J. and J. S. Rovner. 1980. Trail-following behaviour in two species of wolf spiders: sensory and etho-ecological concomitants. Anim. Behav., 28:735-741. Whitcomb, W. H. and R. Eason. 1965. The mating behavior of Peucetia viridans (Araneae: Oxyopidae). Florida EntomoL, 48:163-167. Manuscript received December 1986 , revised May 1987 . Paz, N. 1988. Ecologia y aspectos del comportamiento en Linothele sp. (Araneae, Dipluridae). J. Arachnol., 16:5-22. ECOLOGIA Y ASPECTOS DEL COMPORTAMIENTO EN LINOTHELE SP. (ARANEAE, DIPLURIDAE) Nicolas Paz S. Departamento de Biologia Universidad de Antioquia Medellin, Colombia S.A. ABSTRACT A species of Linothele (Dipluridae), was studied in Soberania Park (Panama) and the lower elevation, more humid Department of Choco (Colombia). The purpose of the study was to determine behavior related to web construction, prey capture, design and trophic efficiency of the web, food source, migration, agonistic behavior and associated kleptoparasites. Differences between temperatures during the day were also studied inside and outside the spider’s retreat. Positive correlations in the variables of body weight vs. body length; weight vs. diameter of the web orifice; body weight vs. the maximum web dimension; body length vs. maximum web diameter were found in both zones. The webs of these spiders are characterized by having a number of associated symbionts, some of them true kleptoparasites. The spiders discriminate in their choice of prey and are most active at night. There was no evidence that they could excavate their own retreat cavities and they showed high levels of inter- and intraspecific agonistic interactions. RESUMEN El trabajo se inicio en el Parque de la Soberania de Panama y se continue en Colombia (Departamento del Choco), siendo esta ultima area mucho mas lluviosa y de menor altitud. Se trabajo con una especie do Linothele (Dipluridae) con el objetivo de estudiar algunos patrones de comportamiento relacionados con su conducta tejedora; manejo de la presa, patron y eficiencia trofica de la tela, posibles fuentes de alimento, capacidad migratoria, conducta agonistica, cleptoparasitismos, ciclos termicos entre las 0800; 1200; 1600; 2000, de la temperatura ambiental y la del interior de la cueva; ad e mas de algunos aspectos ecologicos. La correlacion de variables (peso x largo corporal; peso corporal x el mayor valor de la dimension de la tela; peso x diametro del orificio de la tela; largo corporal x diametro y por mayor valor de la tela), evidenciaron en las dos zonas de estudios, coeficientes positivos. Ademas, las telas de estas aranas se caracterizaron por presentar un buen numero de simbiontes asociados, algunos ocasionales y otros verdaderos cleptoparasitos. Se encontro que discriminaban presas y no dieron evidencia que contruyen sus propias cuevas; su mayor actividad es nocturna y presenta un alto grado de conducta agonistica intra e interespecifica. INTRODUCCION Actualmente existe en Colombia un gran vacio cientifico relacionado con las investigaciones en los diversos aspectos biologicos de nuestra aracnofauna, con excepcion de los esporadicos trabajos de tipo taxonomico realizados por misiones extranjeras. Ante tal situacion, y luego del conocimiento derivado de la 6 THE JOURNAL OF ARACHNOLOGY investigation en el Departamento de Antioquia, Paz (1978), se diseno este trabajo con el fin de obtener information de aspectos biologicos de una especie del genero Linothele presente en bosques de Panama y Colombia, tales como: patrones de comportamiento relacionados con la defensa, captura de la presa, interacciones agonisticas, construction de redes y eficiencia de las mismas, posibles causas de muerte y migration, discrimination de presas, cleptoparasitos y otros simbiontes asociados con la tela, ciclos de temperatura en las cuevas, actividad diurna y nocturna y hasta donde fuera posible, aspectos de su biologia reproductiva. Las observaciones se harian sin descuidar las del laboratorio. Interacciones entre cleptoparasitos y huesped en arenas constructoras de telas aereas ban sido descritas por Thornhill (1975), Vollrath (1978-1979b), Turnbull (1964), Rypstra (1981), Opell y Eberhard (1984); la importancia de las vibraciones inducidas a la tela en la comunicacion intra e interespecifica en muchos grupos de aranas, ha sido estudiada o revisada por Walcott (1959), IJetz y Stratton (1983), Parry (1965) y Vollrath (1979a). Robinson and Robinson (1980) han estudiado el afecto de la captura de la presa y de la destruction de telas orbitales en varies grupos de aranas, lo mismo que su conducta durante el acto de captura de la presa, cortejo, apareamiento, y el comportamiento de construction de telas. MATERIALES Y METODOS La investigation se realize en dos areas biogeograficas diferentes: en el Parque de la Soberania de Panama y en el Department© del Choco (Colombia). En Panama, (Fig. 1), se trabajo a traves de la carretera que desde Gamboa conduce al antiguo oleoducto del Darien (Pipeline-Road), cruzando un extenso bosque primario muy bien conservado, con alturas sobre el nivel del mar comprendidas entre los 180 y 250 m, una temperatura promedia de 26.8° C (max = 30°; min — 24.4°) y una humedad relativa promedia annual de ± 80%. En este bosque primario seco, los nidos de aranas predominan a nivel de borde de quebradas principalmente. El area de estudio en Colombia (Fig. 2), corresponds a un vasto sector comprendido entre Tutunendo — Quibdo y Yuto, sitios de facil exploration aun durante horas nocturnas. Alii las condiciones climaticas son mucho mas inestables que en el sector de Panama, con alturas sobre el nivel del mar comprendidas entre los 40 y los 56 m, temperatura promedia diaria con maximas y minimas similares al area de Panama y una humedad relativa promedio anual entre 86 y 96%. En ambas areas, luego de su respective reconocimiento fisiografico, se seleccionaron los sitios de trabajo en concordancia con la abundancia de nidos y la facilidad de transit©. Asi, se procedio a marcar (con cinta roja) y a medir las dimensiones de las telas visibles. Las medidas se tomaban a partir del orificio de entrada de la cueva, frontalmente hacia el observador (L = largo) y en sentido transversal pasando por el orificio de la tela (A = ancho), ademas se determinaba el valor del diametro del orificio de entrada a la cueva. Si la ubicacion de la tela lo permitia, se procedia a tratar de capturar la arana, induciendo su salida con presas de artropodos vivas o con umbrales de vibraciones artificiales producidas con una varilla delgada o bien cavando con una barra o pala pequena. Si la arana se capturaba se media su longitud desde la PAZ — ECOLOGIA Y COMPORTAMIENTO EN LINOTHELE 7 insertion de los queliceros hasta el tuberculo anal y luego se le pesaba viva encerrada dentro de un pequeno recipiente plastico con una balanza Ohaus modelo — 700. La arana podia traerse al laboratorio para marcar su cefalotorax, con liquido corrector para escritura a maquina o vinilo y regresarla a un sitio distinto o al de captura, bien sobre la tela o en la cueva, para determinar su capacidad de residencia en el lugar (o de migration en caso de encontrarla posteriormente en nuevas telas o sitios). Al cavar los nidos se procuro seguir el patron arquitectonico del tunel para apreciar su complejidad (Figs. 3 a 6). Algunas aranas marcadas se trasladaron de un lugar a otro, donde se soltaban o se colocaban en depresiones naturales o hechas por nosotros con el fin de 8 THE JOURNAL OF ARACHNOLOGY determinar si las mismas cavaban su propia cueva tejiendo luego la tela; o si aprovechaban las depresiones naturales existentes. Lo anterior, porque durante las excavaciones se encontraron pequennos monticulos de tierra removida al parecer por la arana. Su mantenimiento en condiciones de cautiverio se llevo a cabo en nuestra microestacion biologica en recipientes de plasticos, en donde recibian el ciclo normal de luz y oscuridad. Con intervalos de cuatro horas, se determino la existencia de diferencias de temperatura en el ambiente de las cuevas y de su medio circundante, procediendose para ello a medirla con un termometro electronic© Cole/ Parmer/ Chicago a las 0800; las 1200; las 1600 y las 2000. La posible actividad de los cleptoparasitos y otros simbiontes asociados, se trato de determinar principalmente en las telas de bosque por su relativa abundancia en relacion con las telas de las areas de borde de carretera. PAZ — ECOLOGIA Y COMPORTAMIENTO EN LINOTHELE 9 Figs. 3-9. — Diagramas correspoedientes a algueos de los perfiles subterraneos de las cuevas y patrones de telas observadas para nidos de Dipluridae; 3-6, patrones de cuevas encontrados en terrenes arcillosos; 7-9, patrones de telas con su orificio central. Las observaeiones de interaceiones agonisticas intra e interespecificas se faicierom capturando ejemplares de Linothele u otros generos y se marcaba la arana considerada intrusa. Para la identification de los ejemplares de ambas areas y de los simbiontes asociados a las telas, se enviaron muestras a especialistas eomo los doctores R. J. Raven; H. W. Levi; B. D. Opell y otros se identificaron con materia! bibliografico tales como: Kastoe (1978); Forster y PI a truck (1977); Exline (1962); Petruekevitch (1925, 1929); Levi (1963, 1967) y Gertsch (1979). RESULTADOS Y DISCUSION En el area de Panama, los nidos de estas aranas se comenzaron a eecontrar a los 4,000 m, de penetration a partir de Gamboa (sobre el canal), con increment© de abundaecia desde rio Sirystes hasta el sector de agua — salud. En Colombia, los nidos predominaron sobre los bordes de carreteras, lo que hizo mas facil el trabajo con las telas marcadas. Las aranas fueron identificadas como Linothele sp. por R. J. Raven (Queensland Museum, Australia), quien consider© que posiblemente pertenecian- a la misma especie, por lo cual trabajaria en su determination final. Los ejemplares de Panama (Fig. 10), se caracterizaron por presentar uea coloration cafe con visos satinados en su cefalotorax, dad a por una tupida red de vellosidades sobre su cefalotorax. Abdominalmente las vellosidades son mas esparcidas y el matiz cromatico que le dan al abdomen es de una coloration mas opaca. Su tela es mas enmarafiada. Los ejemplares de Colombia, presentan una coloration somatica mas criptica, con la coloration del cefalotorax mas opaco, con visos entre verde, ocre y con una marcada margieation colateral cafe. Su tela ligerameete meeos enmarafiada, Considerando que son de la misma especie, es posible que las difereecias fenetieas entre los dos grapes se debae a factores edaficos y ecologicos y eo a que sean especies taxonomicamente diferentes. 10 THE JOURNAL OF ARACHNOLOGY Fig. 10. — Hembra de un ejemplar de Linothele gravida. Nido y capacidad territorial. — De nuestras observaciones se puede evidenciar que las aranas de ambas areas no siguen un patron secuencial durante la construccion de su complicada tela sobre alguna depresion y tampoco presentan un patron geometrico fijo detectable en algunas telas de Araneidae las cuales son principalmente aereas. A pesar de que la construccion y reparacion de la tela puede realizarse en cualquier momento, siempre y cuando no existan factores de perturbation, la mayor actividad constructora se realiza durante la noche, periodo en donde se puede observar a la arana separarse incluso varios metros de la boca del tunel para ubicar los hilos de soporte en extremos opuestos. Durante la reparacion o construccion, el desplazamiento de la arana es direccionalmente irregular, hacia adelante, atras, o a los lados, y con giros abdominales colaterales durante los cuales su par de largas hileras posteriores suelen cambiar de angulos normalmente con relation al eje hipotetico antero-posterior del cuerpo y entre ellas mismas, al acercarse o separarse. Mientras el pequeno par anterior parece producir un tipo de seda que actua como cemento para pegar entre si los hilos que conforman la estructura de soporte de la tela y para pegarlos sobre sustratos (hojas, piedras, ramas, tallos, etc.). Los multiples, transparentes y finos filamentos de seda que emergen de sus hileras posteriores (principalmente del ultimo o tercer segmento), son los responsables de ocasionar el mayor grado de enmaranamiento (Fig. 11). La tela suele presentar normalmente varios estratos de filamentos de seda a menudo tan bien fusionados que son capaces de retener gotas de agua lo que facilita su visualization durante o despues de periodos de lluvia. La boca del tunel queda normalmente en el centro de la asimetrica tela (Figs. 7-9) continuandose con la depresion a traves de hilos mediante los cuales la arana, desde el interior de la cueva puede comunicarse con el exterior de la plataforma. La longitud de la depresion suele variar de acuerdo con la naturaleza del terreno y la fisiografia del area, pues las aranas pueden aprovechar grietas externamente pequenas pero complicadas en el interior, como en ciertos casos de areas de raices de arboles o fallas rocosas. PAZ — ECOLOGIA Y COMPORTAMIENTO EN LINOTHELE 11 Fig. Ill — Una de las hileras posteriores del abdomen de Linothele sp. con filamentos de seda proyectandose desde sus tubuli. Durante el proceso de construccion o reparacion, las aranas suelen acicalarse sus tarsos, metatarsus y en ocasiones desde la patela y tibia cuyos miembros son introducidos doblados y tirades hacia atras eetre los queliceros. Cuando se procedia a destruir el nido y su tunel, con el fin de capturar la arana, se observaron ocasionalmente sectores modificados con tierra pulverizada en el piso, lo que nos llevo a pensar que las aranas posiblemente podian intervenir activamente en la fabricacion del tunel, de manera especial cuando el terreno era arcilloso y blando. Para corroborarlo se transladaron 44 ejemplares marcados (incluidas ambas areas de estudio), de su sitio original a ueo nuevo, en donde se les colocaban adyacentes a orificios de depresiones naturales o artificiales (cavadas por nosotros). Revisadas desde el dia siguiente durante un mes, se encontro que 13 (29.4%), de estos nuevos sitios fueron abandonados, no asi el resto (31 = 70.6%). En ambas situaciones, las aranas no dieron evidencias que pudieran construir sus propios nidos cavando a traves de sustrato blando, aunque en 3 6 4 de los huecos artificiales se encontro algo de tierra pulverizada, cuyo origen no fue posible dilucidarlo. En los 13 nidos abandonados se incluyeron cuatro que no fueron encontrados (perdidos), al parecer por derrumbe. Esta situacion ademas de factores como: area pobre en hojarasca, nido muy expuesto al sol, huecos inundables, pobreza de humedad y predacion de roedores u otros, fueron posibles causas del abandono del nuevo sitio. Esta ultima situacion no parece ser muy valida para Colombia, pero si para Panama, en donde de acuerdo con M. H. Robinson (comentario personal), el ha observado en la isla de Barro Colorado a los mamiferos Nasua nasua (gato solo; cosumbo solo; coatimundi), cavar y alimentarse de tales aranas y de otras como Theraphosidae, que habitan en depresiones subterraneas. En condiciones de cautiverio, tres de las aranas capturadas se comieron su extensa tela en menos de cinco dias y no coestruyeron nuevamente, lo que evidencia que esta situacion tambien puede presentarse en su ambito natural aunque se ignoran las causas de esta conducta nidofagica. Algunas aranas capturadas y colocadas en cautiverio construyeron o iniciaron la construccion de su nido inmediatamente, otras solo lo hicieron despues de varios dias o semanas y otras no lo hicieron. Esta conducta parece estar relacionada con estados de gravidez y con el tipo de material presente en el recipiente Porque en donde se agregaba buena hojarasca con troncos viejos (a los 12 THE JOURNAL OF ARACHNOLOGY a .a 4) T1 -a « « S c 2 ■3 * § ii 31^ '3 -r -a «s > « 3 o T3 2 11 3 *J g .2 .Q § s e3 c <3 « .a > x> 3 e © o < e U < u a. PAZ — ECOLOGIA Y COMPORTAMIENTO EN LINOTHELE 13 Fig. 12. — Valores de la temperatura promedio ( X ) y su correspondiente d.t. entre las cuevas de telas de aranas y su medio circundante, para las areas del “Parque de la soberania de Panama” y del Choco, tomadas entre las 0800 y 2000. que se les habia hecho depresiones), ramitas secas y la humedad necesaria, la construction del nido se inicia mas rapidamente. Algunas aranas capturadas y liberadas de nuevo en sitios diferentes mostraron una conducta muy activa en la busqueda de depresiones terrestres, pues fuera de su nido parecen ser facil presa de potenciales predadoras, o pueden morir pronto por action del calor, como se pudo comprobar aun en los recipientes de coleccion. El 80% de las telas marcadas en ambas areas de estudios permanecio con su arana durante los dos meses del estudio en Panama y entre los cuatro y seis meses en Colombia. La tela construida por el macho suele ser mucho mas pequena y en sitios mas solitarios. Ciclos termieos en las cuevas. — A pesar de que esta variable no evidencia afectar los aspectos del compartamiento estudiado, fue considerada para determinar posibles variaciones termicas entre el ambiente de las cuevas y su entorno. Tabla 1 y Fig. 12 muestran que los valores promedios para el ambiente de las cuevas fue mas baja que la ambiental, excepto para las 2000. Ademas, que el area de Panama presenta un mayor grade termico en relation a la de Colombia, lo que posiblemente se debe a las diferencias en precipitation y por ende de humedad relativa, responsables de que las condiciones climaticas en esta area de Colombia sean mas inestables, lo que es reflejado aun en su desviacion tipica (d.t.). La mayor temperatura de la cueva 2000, en ambas areas, se deberia a la lenta radiation del calor absorbido durante el dia por el substracto terrestre. Relation de variables entre aranas vs tela. — La Tabla 2 muestra los valroes mmimos, maximos, promedios y las d.t. del peso y largo corporal de las aranas (PC. x L.C.); del diametro del orificio (D.O.) y el mayor valor del tamano de la tela (M.V.T.) para ambas areas, en donde los valores son normalmente mayores 14 THE JOURNAL OF ARACHNOLOGY Tabla 2. — Representation de los valores mmimos (MIN), maximos (MAX), promedio {X) y sus desviaciones tipicas (d.t.), de las variables correspondientes al peso y al largo corporal, el diametro del orificio de la tela y el mayor valor de la tela, tornado entre su largo y su ancho para las dos areas de estudio. P.C. (g) L.C.(cm) D.O.(cm) M.V.R.(cm) Area MIN MAX X d.t. MIN MAX X d.t. MIN MAX X d.t. MIN MAX X d.t. Panama (n — 36) 0.42 3.2 1.53 0.78 1.42 3.5 2.8 0.64 2.4 5.5 4.1 0.78 35 110 58.2 18.5 Colombia (n = 60) 0.3 3.4 2.1 0,71 0,5 3.6 2,9 0.65 0.4 5.6 4,1 0,92 18.2 105 63.9 18.4 para Colombia. Las Tablas 3 y 4 muestran los valores del coeficiente de correlacion, del t. calculados y los grades de significancia para los parametros anteriores. Observese sobre los mismos que todos los valores de correlacion a pesar de ser positives no son altos con excepcion de la combinacion L.C. x P.C. equivalente a 0.984 (98.4%), para Panama y 0.735 (73.5%), para Colombia. Sobre las matrices, se puede notar que los valores de correlacion fueron mas altos para Panama, excepto para P.C. x M.V.T. y L.C. x M.V.T. que fueron mayores para Colombia, lo que posiblemente se debe a las diferencias de las condiciones ambientales reinantes en las areas de estudio. Cleptoparasitos y otros simbiontes en la tela. — Los cleptoparasitos propiamente dichos se encontraron asociados principalmente a hilos o en diminutas telas adyacentes a la de Linothele a donde se desplazan sobre su plataforma horizontal, bien al sentir vibraciones correspondientes a presas o residues que fueran capturados por la arana residente. Cuando un objeto no vivo o bien algun animal vivo caian bruscamente sobre la tela o producian algun tipo de vibracion anormal, se pudo observar que los cleptoparasitos abandonan rapidamente la plataforma en busca de escondite. Los simbiontes encontrados en ambas areas sobre la tela fue el siguiente: Uloboridae: Philoponeila vittata (Keyserling), P. republicana (Simon); P. tingena (Chamberlin and Ivie); Ariston sp. Araneidae: Cyclosa sp. Theridiidae: Argyrodes atopus (Chamberlin and Ivie), A. cordillera (Exline). Theridiosomatidae: Theridiosoma sp. Tetragnathidae: Mecynometa sp; Leucauge venusta (Walckenaer) Pholcidae: Psilochorus sp. Tabla 3. — Valores de la matrix de correlacion para los parametros peso y largo corporal; diametro del orificio de la tela y mayor valor de la tela, tornado entre su largo y su ancho. P = Significancia para el valor de t calculado. Para area Colombia. n = 60 P.C P L.C. P D.O P M.V.t. P P.C 0.735 0.323 0.387 t = 8.28 <0.005 t = 2.6 <0.01 t = 3.2 <0.005 L.C — — 0.497 0.523 t = 4.4 <0.005 t = 4.7 <0.005 D.O — 0.353 t = 2.9 <0.005 Tela — — PAZ — ECOLOGIA Y COMPORTAMIENTO EN LINOTHELE 15 Tabla 4. — Valores de la matrix de correlation para los parametros peso y largo corporal; diamentro del orificio de la tela y mayor valor de la tela tornado entre su largo y su ancho. N = 36. P = Significancia para el valor de t calculado. Para area Panama. P.C P L.C P D.O P M.V.t. P PC — — 0.984 0.743 0.244 t = 17.37 <0.005 t = 6.5 <0.005 t = 1.47 N.S. L.C — 0.728 0.286 t = 6.2 <0.005 t = 1.74 <0.005 D.O — — 0.552 t = 3.86 <0.005 Tela — — Mysmenidae: Mysmenopsis dipluramigo (Platnick and Shadab); Mysmenopsis sp- Symphytognathidae: Curimagua sp. Thomisidae: Misumenoides sp. Salticidae: Lyssomanes sp. Algunos ejemplares de Lycosidae, Salticidae y Linyphiidae no fueron identificados. Las familias mejor representadas correspondieron a Ulboridae y Theridiidae por su predominante presencia en ambas areas, pero especialmente a nivel de bosque, las especies de P. republicana seguida de A. atopus consideradas como verdaderas cleptoparasitas en la tela de Linothele. A muchas especies o generos de otras familias se les consider© simbioetes ocasionales ya que podian aprovechar la tela para capturar pequenas fuentes troficas que pasarian desapercibidas para las Dipluridae. Vollrath (1978, 1979a, 1979b) y Rypstra (1981), estudiando la posible funcion de los cleptoparasitos en la tela de ciertas aranas tejedoras encontraron que fuera de limpiarla de detritos o presas dejadas por la arana huesped, eran capaces de diferenciar tipos de vibraciones producidas por esta durante un acto de captura de presa, monitorear sus movimientos sobre la tela y asi aprovechar sus descuidos para hurtarse algunas pequenas presas inactivas. Como Walcott (1959) lo demostro, la capacidad de ubicar la presa sobre la tela u otros agentes inductores de vibraciones, es practicamente dependiente de sus receptores tarsales sin los cuales aun la capacidad de discriminar sera afectada y de alii la importancia del acicalamiento de estos segmentos de sus miembros. Otros organismos fueron considerados como asociados ocasionales, al encontraseles con muy baja frecuencia en algunas de las telas. Tales grupos fueron: Falangida, Hemiptera, Coleotera, Quilopoda y larvas de Lepidoptera. De estos, fue casi constante la presencia de un diminuto hemiptero de la familia Reduviidae Metapterus sp., caracterizado por su gran semejanza con los fasmatides, por lo cual se confuede con esos Orthoptera. Fuentes troficas. — De acuerdo con nuestras observaciones directas e indirectas a traves de restos de alimento, las fuentes troficas de estas aranas suelen ser muy variadas. Asi, tenemos las siguientes bien en su fase adulta o larvaria: Orthoptera adultos (Blattidae y Grillidae), Hemiptera adultos, Hymenoptera adultos, Lepidoptera (principalmente larvas), pequeiios Coleoptera, Isopoda (Crustacea), Diplopoda, otras aranas y aun pequeiios saurios (largartija) y anuros (sapos y ranas). De estos grupos las presas mas abundantes encontradas durante las horas 16 THE JOURNAL OF ARACHNOLOGY nocturnas corresponds a ejemplares de Blattidae, Grillus , larvas de Lepidoptera. Hemiptera e Hymenoptera adultos nocturnes. Comportamiento de captura de la presa. — El comportamiento de captura de la presa suele variar y por ende las unidades de secuencia del etograma con marcadas difereecias depeediendo del tipo de la fuente trofica, por lo cual se deduce que estas ar arias suelen discriminar presas vivas en concordancia con el patron de vibraciones que produzcan como ha side demonstrado por Parry (1965) y Vo I) rath (1979a), o induciendo estas con artefactos meeanicos. Para su corroboration, se experiment© con varies tip os de presas a diferentes horas, entre las 0800 y las 2000, cada dla de campo, dedicando el tiempo que fuese necesario de observation para cada uno de los 25 ensayos h echos con cada presa. Estas diferencias de coeducta durante la captura de presas, se pueden observar en las Figs. 13-16 donde la frecuencia de las unidades ee relation a su secuencia suele variar y el espesor de los vectores en los etogramas, representa la mayor o menor frecuencia de repetition de las unidades de actividad. La Fig. 13 esta relacionada con la captura de un Hymenoptera (avispa), en donde la arana trata de inactivar la presa con una mordida prolongada y cuando la suelta, la toca con sus miembros anteriores suavemente, mordiendola de nuevo si la detecta viva antes de iniciar su digestion. Esto podria atribuirse a que la arana pareciera reconocer que la presa presenta un mecanismo defensive (aguijon venenoso) altameete efectivo y trata de evitar sus consecueecias. Las Figs. 14 y 15, ambas pertenecientes a presas de Grthoptera, se caracterizan por su mayor complejidad en unidades de secuencias de actividad. En ambas, igual que ee las Figs. 13 y 16 notese que la conducta de inactivar la presa envolviendola primero no se presenta, a pesar que en algunos casos se observe que luego de morder, trata de envolverlas. La Fig. 15 muestra un mayor numero de unidades de secuencias, que diferencian este etograma del 14, posiblemente debido a la conducta de la arana frente a una presa con gran numero de procesos espinosos sobre sus pa las, dispositivos que podrian lesioear al delicado tegument© de la arana. For esta razon posiblemente tambien ocurre su inactivation tratando de envolverlo en tela, lo que es poco comun durante la captura de presa en represeetantes de la familia Dipluridae. La Fig. 16 esta relacionada con la captura de una presa que se caracteriza por tener un exosqueleto altamente queratinizado y calcificado (diplopodos), lo que podia ser una de las posibles causas de que estos artropodos no scan edibles presas para las aranas, complementado por la action de las gland ni as repugeatorias de los mismos. El tiempo de ingestion y digestion suele variar de acuerdo con el tamafio y naturaleza de la presa, del estado fisiologico de la arana y de si la captura se hace sobre una tela terminada o en construction. No es normal encontrar en el interior de la cueva o sobre la tela, restos de presas ya que estas aranas sacae sus residues al terminal el proceso ingestivo y digestive, cuyas partes blaedas aun presen res, suelen ser aprovechadas por otros organismos, especialmente hormigas. Actividad agonistica. — La interaction agoeistica se maeifiesta con aranas del mismo genero y diferentes especies, por ejemplo; si una Linothele sp. consider ad a como intrusa se coloca sobre la tela de otra (residente), la primera puede quedarse inmovil inicialmeete para luego fauir, o huir inmediatamente al ser PAZ — ECOLOGIA Y COMPORTAMIENTO EN LINOTHELE 17 Fig. 13. — Patron de secuencia de unidades del comportamiento de captura de un Hymenoptera (Avispa). El espesor de los vectores iedica la frecuencia con que se repitio un a actividad. perseguida per ia residente, la que en caso de capturarla sera predada. Es de resaltar que aunque ia intrasa sea de mayor tamano que la residente, no suele manifestar agresividad a esta ultima, cosa que si exhibe ia residente. Este patron de comportamiento se pudo observar con intrusas de diferentes geeeros colocandolos con la mano sobre la red de la residente y en terrarios tales comoi Argiope , Nephiila , Gasteracantha , Leucauge , Micrathena , Lycosa y otras Dipluri- dae. Si en condiciones de laboratorio se juntan dos aranas de igual tamano en un tena.no, estas exhiben acciones agonisticas entre si y cualquiera de las dos podia ser la victima. Pero ee caso de que una de ellas inicie la construccion de su nido primero que la otra, aquella por lo general se torna mas agresiva y normalmente 18 THE JOURNAL OF ARACHNOLOGY Fig. 14. — Patron de secuencia de unidades del comportamiento de captura de una Blattidae ( Periplaneta americana), por Linothele sp. El espesor de los vectores indica la frecuencia con que se repitio una actividad. mata a la companera. Si hay diferencia de madurez sexual la arana mas adulta normalmente mata a la mas joven. Sin embargo, esta conducta puede variar si se juntan machos y hembras inmaduros o hembras entre si inmaduras, en donde ambos ejemplares manifiestan actividad agonistica sin llegar al acto de predacion pudiendo convivir durante mucho tiempo. Esta situacion tambien se modifica si a una hembra sexualmente madura pero no receptiva se le agrega un macho inmaduro o adulto, En ambos casos suele devorarlos, no asi si la hembra esta receptiva y el macho sexualmente maduro; en este ultimo caso hay acciones agonisticas dentro de los movimientos de galanteo PAZ — ECOLOGIA Y COMPORTAMIENTO EN LINOTHELE 19 Fig. 15. — Patron de secuencia de unidades del comportamiento de captura de un Orthoptera (Grilles), por Linothele sp. El espesor de los vectores iedica la frecuencia con que se repit) 6 una aetividad. precopuiatorio pudiendo coevivir en el mismo recipiente por varios dias con el macho aunque este carezca de espermatoforos en sus palpos. La agresividad entre estas arafias es muy posible que dependa del numero de mud as padecidas por los ejemplares ya que mientras mas inmaduras eran las aranas meeor fue la expresion de agresividad entre ellas y mientras mas proxima a su madurez sexual mayor era esta expresion agoeistica, lo cual espero corroborar en la segunda fase del estudio relacionado solo con la biologxa reproductiva y desarrollo postembrionario de las mismas. De euestras observaciones con estas Dipluridae se puede deducir que las mismas no contruyee realmente sus cuevas, como si lo hacen otras familias del 20 THE JOURNAL OF ARACHNOLOGY Fig. 16. — Patron de secuencia de unidades del comportamiento de captura de un Diplopoda carinado ( Polidesmus sp.). El espesor de los vectores indica la frecuencia con que se re pit;. 6 una actividad. ordee aunque existe la posibilidad de que si el terreno lo permite modifique algunos seetores de la depresioe natural, lo que representaria para estas especies una gran veetaja desde el pueto de vista de costo energetic© en relation a aquellas que si construyen su depresion como en Ctenizidae (Coyle 1981) Lycosidae, Theraphosidae y algunos de Dipluridae del area australiana (Main 1975, 1982). La pobre capacidad migratoria de estas aranas favorece algunas hipotesis de la funcioe de ser territorial concebida por algunos autores y de manera especial en la revision del concept© por Verner (1977) tales como aprovechar recursos y material de nidation, impedir el contagio de enfermedades, aumentar la posibilidad de neutraiizar acciones predadoras, de selection de pareja o de tener exito reproductive y disminuir las ieteracciones agonisticas intra e interes- pecificas. Sin embargo, de acuerdo con lo observado, la riqueza en fuentes troficas no parece ser ueo de los parametros vitales del gran poder de residencia de estas aranas ya que se encoetraron muefaas telas en sitios bastante pobres en fuentes de alimentos aun durante la eoche y ademas, como se pudo demostrar en condiciones de cautiverio (no alimentadas por mas de dos meses), estas aranas suelen soportar largos periodos sin comer, especialmente en la epoca de muda. Por lo tanto, la escasa abundancia de presas tampoco seria causa de abandon© de las telas. PAZ — ECOLOGIA Y COMPORTAMIENTO EN LINOTHELE 21 AGRADECIMIENTOS Doy un reconocimiento especial al Dr. M. H. Robinson y a las autoridades directivas del Smithsonian Tropical Research Institute (Panama) por su invaluable asesoria cientifica, colaboracion bibliografica y demas servicios prestados durante mi permanencia en Panama. Agradezco al Dr. R. J. Raven del CSIRO (Division de Entomologia del Museo Nacional de Australia), por su colaboracion en la identificacion de los ejemplares encontrados en Choco (Colombia) y en Panama. Igualmente a los doctores IT W. Levi (Museum of Comparative Zoology, Harvard University); F. A. Coyle (Western Carolina University) y B. D. Opell (Virginia Polytechnic Institute and State University), por su colaboracion en la indentificacion y el suministro de bibliografia de los simbiontes encontrados en las telas de las Linothele; a los Drs. N. I. Platnick (American Museum of Natural History); Maria E. Galiano (Museo Hist. Natural Argentina); C. E. Valerio (Universidad de Costa Rica), por su colaboracion en la correction inicial del manuscrito; al Centro de Computos del Departmento de Ingenieria de Sistemas de la Universidad de Antioquia por su asesoria en procesamiento de datos; al Comite Central de Investigaciones y al Centro de Investigaciones de la Facultad de Ciencias Exactas y Naturales por haberme proporcionado todos los medios necesarios para realizar el proyecto; a la profesora del Departmento de Biologia, Maria Luisa Bravo, por su asistencia en el analisis estadistico de los datos y al Biologo Hernando Pareja Mesa, por su colaboracion como asistente de campo durante todo el trabajo. Finalmente deseo agradecer al Dr. William B. Peck, Editor asociado del Journal (Missouri State University) por su acertada revision final. LITERATURA CITADA Coyle, F. A. 1981. Notes on the behaviour of Ummidia trapdoor spiders (Araneae; Ctenizidae): burrow construction, prey capture, and the functional morphology of the peculiar third tibia. Bull. British Arachnol. Soc., 5(4): 159-165. Exline, H. and H W. Levi. 1962. American spiders of the genus Argyrodes (Theridiidae). Bull. Mus. Comp. Zool., 127(2):75-204. Forster, R. R. and N. I. Platnick. 1977. A review of the spider family Symphytognathidae (Araneae: Arachnida). Amer. Mus. Novitates, 2619:1-29. Gertsch, W. J. 1979. American Spiders. 2nd ed. Van Nostrand-Reinhold. 274 pp. Kaston, B. J. 1978. How to Know the Spiders. 3rd ed. Wm. C. Brown. Dubuque. 272 pp. Levi, H. W. 1963. American spiders of genus Theridion (Araneae, Theridiidae). Bull. Mus. Comp. Zool., 1 29( 1 0):483-59 1 . Levi, H. W. 1967. Habitat observations, records and new South American theridiid spiders (Araneae: Theridiidae). Bull. Mus. Comp. Zool, 132(2):2 1-37. Main, B. Y. 1975. The citrine spider: A new genus of trapdoor spider (Mygalomorphae; Dipluridae). Western Australian Nat., 13(4):73-78. Main, B. 1982. Adaptation to arid habitats by mygalomorph spiders. Evol. Flora. Fauna Australia. No. 53. Opell, B. D. and W. G. Eberhard. 1984. Resting posture of orb-weaving uloborid spiders (Araneae, Uloboridae). J. Arachnol, 1 1(3):297-307. Opell, B. D. 1979. Revision of the genera and tropical American species of the spider family Uloboridae. Bull. Mus. Comp. Zool., 148(10):443-549. Parry, D. A. 1965. The signal generated by an insect in spider’s web. J. Exp. Biol, 43:185-192. Paz, S. N. 1978. Introduction a la artrofauna del Depto de Antioquia. Actualidades Biol, 7(23):2- 13. Petrunkevitch, A. 1925. Arachnida from Panama. Trans. Connecticut. Acad. Art. Sci., 27:51-248. Petrunkevitch, A. 1929. Spiders of Porto Rico. Trans. Connecticut Acad. Art. Sci., 30(1): 1-158. 22 THE JOURNAL OF ARACHNOLOGY Robinson, M. H. 1971. Units of behaviour and complex sequence in predatory behaviour Argiope argentata (Fabricius) (Araeeae: Araneidae). Smithsonian Contributions Zool. 65:1-35. Robinson, M. H. and B. Robinson. 1980. Comparative studies of the courtship and mating behaviour of tropical araneid spiders. Pacific Insect Monogr. 36. Rypstra, L. 1981. The effects of kleptoparasites on prey consumption and web relocation in a Peruvian population of the spider Nephila clavipes. Oikos, 37:179-182. Thornhill, R. 1975. Scorpionflies kleptoparasites of web-building spiders. Nature, 28:709-711. Turnbull, A. L. 1964. .The search for prey by a web-building spider ( Archaearanea tepidariorum ) (Theridiidae). Canadian Entomoh, 64:568-569. Uetz, W. G. and G. E. Stratton. 1983. Communication in spiders. Endeavour, 7(1): 13-18. Verner, J. 1977. On the adaptive significance of territoriality. Amer. Natur,, 1 1 1(980):769-775. Vollrath, F. 1978. A close relationship between two spiders (Arachnida Araneidae): Curimagua bayano synecious on a Diplura species. Psyche, 85(4):347-353. Vollrath, F. 1979a. Vibrations: their signal function for a spider kleptoparasite. Science, 205:1149- 1151. Vollrath, F. 1979b. Behaviour of the kleptoparasite spider Argyrodes eievatus (Araneae; Theridiidae). Anim. Behav,, 27:515-521. Walcott, C. and W. G. Verner Kloot. 1959. The physiology of the spider vibration receptor. J. Exp. Zool, 141(2): 191-243. Manuscript received March 1986, revised June 1987. Acosta, L. E. 1988. Contribucion al conocimiento taxonomico del genero Urophonius Pocock, 1893 (Scorpiones, Bothriuridae). J. Arachnol., 16:23-33. CONTRIBUCION AL CONOCIMIENTO TAXONOMICO DEL GENERO UROPHONIUS POCOCK, 1893 (SCORPIONES, BOTHRIURIDAE)1 Luis Eduardo Acosta Catedra de Zoologia I Facultad de Ciencias Exactas, Fisicas y Naturales Universidad Nacional de Cordoba Casilla de Correos 122, 5000 Cordoba, Argentina ABSTRACT In this paper, three species-groups within the genus Urophonius are recognized: exochus group, granulatus group and brachycentrus group. The diagnostic characters used at this level are: arrangement of ventral submedian carinae in metasomal segment I; ventral chaetotaxy in metasomal segments I, II and III; telotarsal spine formula (legs III and IV); morphology of lobe region of hemispermatophore; relative location of femur trichobothria d and e , and macrosetae Ml and M2; ventral pigmentation pattern of metasoma. A key for the nominate species in the genus, as well as some comments on distribution and bioecology, are added. RESUMEN Se reconoce la existencia de tres grupos de especies dentro del genero Urophonius : grupo exochus , grupo granulatus y grupo brachycentrus. Los caracteres diagnostics empleados a este nivel son: disposition de carenas ventrales submedianas en segmento caudal I; quetotaxia ventral en segmentos caudales I, II y III; formula de espinulacion telotarsal (patas III y IV); morfologia de la region de lobulos del hemiespermatoforo; position relativa de las tricobotrias femorales d y e, y las macroquetas Ml y M2; patron de pigmentation ventral en metasoma. Se incluyen una clave para las especies nominadas del genero, asi corno algunos comentarios sobre distribution y bioecologia. INTRODUCCION Entre los generos de Bothriuridae mas dificiles para su estudio taxonomico se encuentra Urophonius Pocock, 1893. Las dificultades de hallar caracteres diagnosticos seguros en el nivel especifico llevaron a frecuentes confusiones de los investigadores, quienes ya a fines del siglo pasado, cuando el genero contaba apenas con tres especies nominadas, tenian problemas para distinguir dos de ellas -U. brachycentrus (Thorell, 1877) y U. jheringii Pocock, 1893, durante largo tiempo considerados como sinonimos-; tambien U. granulatus Pocock, 1898 fue motivo de controversias, ya que, como lo senala Maury (1979), “ha pasado por lo menos bajo 5 denominaciones distintas” El posterior aumento del numero de 'Trabajo realizado en el Museo Argentine de Ciencias Naturales (Buenos Aires); presentado en las II Jornadas Cientificas de la Sociedad de Biologia de Cordoba, 1985. 24 THE JOURNAL OF ARACHNOLOGY especies, en algunos casos injustificado, complied el panorama, en tanto similar efeeto tuvo un articulo de San Martin (1965), en el que, tras proponer acertadamente la sinonimia generica de Urophonius con lophorus Penther, 1913, designa a ejemplares de U. granulatus como “alotipo” y “paratipo macho” de U. eugenicus (Mello-Leitao, 1931) (Maury,- 1979). Un primer intento de aclarar la situation correspoede a Maury (1973), que postula la subdivision del genero en dos “grupos de especies”, basandose principalmente en la formula de espinulacion telotarsal. Entiendo que tal division es valida parcialmente, pues, si bien el denominado “grupo A” incluye especies claramente afines (equivale al concepto de lophorus ), no ocurre lo mismo con el “grupo B” ee el cual pueden distieguirse al menos dos grupos diferentes. En un seguedo aporte, sin volver a referirse a tales grupos, Maury (1977) menciona una serie de caracteres que considera adecuados para la identification de las especies. Dichos caracteres -y otros que deben anadirse-- pueden ser jerarquizados, siendo algunos de ell os utiles para distinguir grupos de especies, como veremos, en tanto otros solo tienee valor a nivel especifico. Luego de estudiar las especies conocidas del genero -except© U. tumbensis Cekalovic, 1981-, he comprobado que es posible reunirlas en ties grupos, a los que he denominado con el nombre de su especie mas caracteristica: grupo brachycentrus , grupo granulatus y grupo exochus (este ultimo coincide con el “grupo A” de Maury, 1973); para su distincion he tornado en cuenta una serie de siete caracteres, a los que me referire en el punto siguiente. La ubicacion de U. tumbensis en alguno de estos grupos, asi como su inclusion en la clave que acorn pa ha este trabajo, quedan pendient.es, por cuanto no he podido exarninar ningun ejemplar de dicha especie, y los datos proporcionados en la description original son insuficientes. Aunque los aqui propuestos son, sin dud a, autenticos grupos naturales -probabiemeete representan sendas line as filogeneticas-, la magnitud de las diferencias no parece ser suficiente para otorgarles categoria subgeneriea. Por de pronto, su reconocimieeto podra servir de orientation, en el esclarecimiento sistematico de un genero tan homogeneo como lo es Urophonius. CARACTERES UTILIZADOS Los grupos de especies de Urophonius pueden reconocerse por los siguientes caracteres: (1) careeas ventrales submedianas del segment© caudal I; (2) disposition de quetas ventrales en el mismo segmento; (3) disposition de dichas quetas en los segmentos caudales II y III; (4) espinulacion telotarsal en patas III y IV (formula mas frecueete); (5) morfologia de la region de lobules en el hemiespermatoforo; (6) position de las tricobotrias d y e en el femur respecto de las macroquetas indicadas como Ml y M2 en Figs. 12 a 14; (7) patron de pigmentation ventral en el metasoma. La nomenclatura que empleo para las carenas caudales responde a lo propuesto por Francke (1977), mientras la correspondiente a los lobulos del hemiespermatoforo es la utilizada por San Martin (1963); las siglas tricobotriales ban sido tomadas de Vachon (1973). En cuanto a las macroquetas del femur referidas en el caracter 6, su denomination es arbitraria, al solo efeeto de su individualization en las figuras. De los caracteres considerados, la morfologia de la region de lobulos del hemiespermatoforo es sin duda el mas unportante, En el genero Urophonius el ACOSTA— TAXONOMIA DE UROPHONIUS 25 Figs. 1-5. — Carenas ventrales submedianas del granulatus ; 2, Urophonius granulatus ; 3, otras es] (Chile central); 5, otras especies. segmento caudal I: 1, grupo exochus; 2-3, grupo ecies; 4-5, grupo brachycentrus; 4, Urophonius sp. hemiespermatoforo responde, en vista externa, a un unico modelo basico, donde pequenas variaciones en la lamina distal y las apofisis representan diferencias especificas (Maury, 1980). Por el contrario, el estudio de la region de lobulos -en vista interna- permite reconocer la existencia de tres patrones bien definidos (correspondientes a cada grupo de especies), segun el distinto desarrollo de los pliegues formados entre el lobulo interno (l.i.) y el lobulo basal (Lb.), y entre este y el lobulo externo (l.e.); es tambien importante la presencia y morfologia de una prolongacion laminar en el extremo del l.b. Es de notar que ninguno de los caracteres restantes puede ser empleado individualmente para una separacion categorica de los tres grupos de especies, y por este motivo deben tomarse necesariamente en su conjunto. Otros caracteres (tales como la forma del telson, el numero de dientes pectineos, los disenos cromaticos de prosoma y tergitos, etc.), subordinados a los que se emplean en este trabajo, o bien estados particulares de estos, permiten la distincion de las especies, pero con diferente valor dentro de cada grupo. El presente articulo incluye una clave para diferenciar las especies nominadas de Urophonius , donde se hace mencion de estos caracteres del nivel especifico. GRUPO EXOCHUS Diagnosis. — Segmento caudal I: carenas ventrales submedianas longitudinales, algo divergentes en su extremo proximal (Fig. 1); tres pares de macroquetas ventrales, siguiendo las carenas (Fig. 6). Segmentos II y III: tres pares de macroquetas (Fig. 6); puede haber un par de microquetas proximales, cercano a los laterales. Espinulacion telotarsal: T III: 4-4, T IV: 4-5. Hemiespermatoforo con la region de lobulos bien desarrollada, particularmente el pliegue formado entre Li. y l.b., mientras el sector entre Lb. y l.e. es apenas una angosta franja (Fig. 9); el lobulo basal carece de prolongacion laminar evidente, pero existe una estructura laminar rudimentaria oculta tras el propio lobulo, quizas equivalente a las prolongaciones presentes en otros grupos. Entre tricobotrias d y e del femur 26 THE JOURNAL OF ARACHNOLOGY Figs. 6-8. — Quetotaxia ventral de los segmentos caudales I a III: granulatus ; 7a, Urophonius treguaiemuensis ; 8, grupo brachycentrus. aparece una sola macroqueta (Ml, Fig. 12), equidistante de ellas. Pigmentacion ventral en metasoma: dos bandas paramedianas irregulares. Especies incluidas y distribution. — Urophonius exochus (Penther, 1913). Argentina: Mendoza, Neuquen, <±Rio Negro? Urophonius eugenicus (Mello-Leitao, 1931), Argentina: Santa Cruz. Urophonius mahuidensis Maury, 1973. Argentina: sur de Buenos Aires, norte de Rio Negro. Urophonius sp.: una forma posiblemente innominada de Perito Moreno (provincia de Santa Cruz, Argentina). Comentarios. — Si bien este grupo es inconfundible como tal, persisten dudas a nivel especifico, por la ausencia de caracteres diacriticos seguros. El estudio del hemiespermatoforo parece confirmar la validez de estas especies, a pesar de su gran semejanza externa. GRUPO GRANULATUS Diagnosis. — Segment© caudal I: carenas ventrales submedianas longitudinales, bien definidas (Fig. 2) o con sus granules disperses (Fig. 3); dos pares de macroquetas ventrales (Fig. 7), menos frecuentemente tres pares, en este ultimo caso puede faltar una queta del par medio (Fig. 7a). Segmentos II y III: tres pares de macroquetas (Fig. 7); pueden agregarse un par de microquetas proximales. Espinulacioe telotarsal: T III: 5-5, 5-6, T IV: 5-6, 6-6 (6-7). Hemiespermatoforo con region de lobulos poco extendida; el l.b. se resuelve en una estructura laminar plana muy evidente, de borde caracteristico (Fig. 10); sector entre l.b. y l.e. concavo. Tricobotria e del femur proxima a la unica macroqueta (Ml, Fig. 13), pero de posicion variable segun la especie. Pigmentacion ventral en metasoma: banda axial y lateroventrales nitidas, paramedianas ausentes. ACOSTA— TAXONOMIA DE UROPHONIUS 27 Figs. 9-11. — Hemiespermatoforo izquierdo, en vista interna: 9, grupo exochus ( Urophonius mahuidensis)\ 10, grupo granulatus (U. granulatus ); 11, grupo brachycentrus ( U . jheringii), l.i. = lobulo interne, Lb. = lobule basal, l.e. = lobulo externo. Especies incluidas y distribution. — Urophonius granulatus Pocock, 1898 = Iophoroxenus exilimanus Mello-Leitao, 1932 = U. paynensis San Martin y Cekalovic, 1968. Argentina: Santa Cruz, Chubut; sur de Chile. Urophonius tregualemuensis Cekalovic, 1981. Chile central. Urophonius sp.: Meseta de Somuncura (provincia de Rio Negro, Argentina). GRUPO BRACHYCENTRUS Diagnosis. — Segmento caudal I: carenas ventrales submedianas longitudinales en su extremo distal, toman una orientacion bruscamente oblicua y divergente en la mitad proximal del segmento (Fig. 4); con frecuencia las ramas oblicuas se conectan en la linea media, separandose de las primitivas carenas longitudinales para formar una unica carena transversal (Fig. 5); granules por lo general perliformes y conspicuos; cuatro pares de macroquetas ventrales, el par proximal sigue la direccion de las carenas, ubicandose cerca de los laterales (Fig. 8). Segmentos II y III: siempre mas de tres pares de macroquetas ventrales (entre cuatro y seis pares), el proximal se acerca a los laterales (Fig. 8). Espinulacion 28 THE JOURNAL OF ARACHNOLOGY Figs. 12-14. — Posicion relativa de tricobotrias d (dorsal) y e (externa), y macroquetas Ml y M2 en femur derecho: 12, grupo exochus\ 13, grupo granulatus\ 14, grupo brachycentrus. En cada grupo, debajo del dibujo correspondiente, se ha graficado en forma esquematica la relacion entre las tricobotrias (circulos blancos) y las macroque- tas (circulos negros); la tricobotria e en el grupo granulatus puede tomar posiciones diferentes, segun la especie. telotarsal: T III: 5-6, 6-6, T IV: 6-6, 6-7. Hemiespermatoforo con region de lobulos bien desarrollada y compleja; la concavidad ubicada entre Lb. y l.e. esta muy extendida, ocultando buena parte del pliegue formado entre el Li. y el propio Lb.; este ultimo termina en una conspicua estructura laminar, de hordes curvados y suavemente concava (Fig. 11). Proxima a la tricobotria d del femur aparece una nueva macroqueta (indicada como M2 en Fig. 14). Pigmentacion ventral en metasoma: un par de bandas paramedianas longitudinales, de borde irregular, que pueden estar acompanadas de un esbozo de banda axial, o confluir en la linea media. Especies incluidas y distribution.^ Urophonius brachycentrus (Thorell, 1877). Argentina; Cordoba, sur de Santiago del Estero, La Rioja, Tucuman, llegando hasta Entre Rios, La Pampa, norte y sur de Buenos Aires. Urophonius jheringii Pocock, 1893 = U. corderoi Mello-Leitao, 1931 = U. granule sis simus Mello-Leitao, 1934. Argentina: sierras de Tandil y Ventana (provincia de Buenos Aires); Uruguay; sur de Brasil. Urophonius achalensis Abalos y Hominal, 1974. Argentina: piso altitudinal superior en las Sierras Grandes (provincia de Cordoba). Urophonius sp.: Chile central. ACOSTA— TAXONOMIA DE UROPHONIUS 29 CLAVE PARA LAS ESPECIES NOMINADAS DE UROPHONIUS 1. Carenas ventrales submedianas del segmento caudal I longitudinales; dos o tres pares de quetas ventrales en dicho segmento, tres pares en los segmentos II y III. Region de lobulos del hemiespermatoforo poco desarrollada, o al menos solo desarrollado el pliegue entre Li. y Lb. Tricobotrias dye del femur relacionadas con una unica macroqueta .2 Carenas ventrales submedianas del segmento caudal I transversales en la mitad proximal del artejo; superficie ventral del mismo segmento con cuatro pares de quetas, siguiendo las carenas, y con mas de tres pares en los segmentos II y III. Region de lobulos muy desarrollada; Lb. con prolongacion laminar concava. Tricobotrias d y e del femur relacionadas con dos macroquetas. Espinulacion telotarsal: T III: 5-6, 6-6, T IV: 6-6, 6-7. Grupo brachycentrus 6 2. Espinulacion telotarsal: T III: 4-4, T IV: 4-5. Region de lobulos del hemiespermatoforo con gran desarrollo del pliegue entre Li. y Lb., y escaso de la concavidad entre Lb. y l.e.; Lb. sin prolongacion laminar. Superficie ventral del metasoma provista de dos bandas de pigmento paramedianas y dos lateroventrales poco definidas, de hordes irregulares. Segmento caudal I con tres pares de quetas ventrales. Tricobotrias dye equidistantes de la unica macroqueta (Ml). Grupo exochus. .3 Espinulacion telotarsal: T III: 5-5, 5-6, T IV: 5-6, 6-6 (6-7). Region de lobulos del hemiespermatoforo pequena; Lb. con prolongacion laminar plana. Superficie ventral del metasoma con dos bandas de pigmento lateroventrales y una axial, bien definidas. Segmento caudal I con dos o tres pares de quetas ventrales. Tricobotria e mas cerca de la macroqueta Ml que la tricobotria d. Grupo granulatus 5 3. Hemiespermatoforo con lamina distal delgada; Li. con protuberancia bifida cercana a la base de la lamina distal. Telson bajo 4 Hemiespermatoforo con lamina distal ancha; protuberancia bifida del Li. alejada de la base de la lamina distal. Telson alto. U. eugenicus 4. Lamina distal del hemiespermatoforo recta; protuberancia bifida unida a la base de la lamina distal U. mahuidensis Lamina distal del hemiespermatoforo suavemente curva; protuberancia bifida sin conexion con la base de la lamina distal. U. exochus 5. Carenas ventrales submedianas del segmento caudal I bien definidas; dos pares de quetas ventrales. Carenas ventrales laterales del segmento caudal V bien definidas. Tricobotria e en posicion mas distal que la macroqueta Ml. Hemiespermatoforo sin repliegue distal posterior, y con lamina distal ancha; Li. con protuberancia bifida en su borde extern© ..... f/. granulatus Carenas ventrales submedianas del segmento caudal I algo dispersas hacia los laterales en una zona granulosa; tres pares de quetas ventrales (puede faltar una del par medio). Carenas ventrales laterales del segmento caudal V poco definidas hacia proximal. Tricobotria e en posicion mas proximal que la macroqueta ML Hemiespermatoforo con repliegue distal posterior y lamina distal delgada; Li. con un denticulo en su cara externa U. . tregualemuensis 6. Carenas ventrales laterales y ventral mediana del segmento caudal V presentes solo en el tercio posterior. Surco interocular suave. Diseno cromatico de la 30 THE JOURNAL OF ARACHNOLOGY superficie ventral del metasoma: manchas paramedianas muy extendidas, por lo general confluyen en la linea media. Macho con pinzas comparativamente mas gruesas. Hemiespermatoforo con lamina distal gruesa, repliegue distal posterior bien desarrollado; l.i. con un par de denticulos en su cara externa. Numero de dientes pectineos: macho, 14 a 15, hembra, 12 a 14. Superficie ventral del segmento caudal III granuloso. Carenas ventrales laterales del segmento I vestigiales. U. jheringii Carenas ventrales laterales y ventral mediana del segmento caudal V mas extendidas. Surco interocular marcado. Dimorfismo sexual en pinzas menos acentuado. Repliegue distal posterior menos desarrollado. Mayor numero de dientes pectineos. Diseno cromatico de la superficie ventral del metasoma: manchas paramedianas mas limitadas, no confluyen en la linea media. Carenas ventrales laterales del segmento caudal I bien desarrolladas 7 7. Tergitos con un par de manchas paramedianas en forma de parentesis. Numero de dientes pectineos: macho, 17 a 20, hembra, 16 a 18. Segmento caudal III granuloso por ventral. Repliegue distal posterior muy pequeno, lamina distal ancha; l.i. con protuberancia bifida en su borde externo .... U. brachycentrus Tergitos con un par de manchas paramedianas subtriangulares. Numero de dientes pectineos: macho, 16 a 18, hembra, 14 a 16. Segmento caudal III liso por ventral (en algunas hembras, escasos granulos). Repliegue distal posterior medianamente desarrollado, lamina distal delgada; l.i. con un par de denticulos en su cara externa. ........................... U. achalensis OBSERVACIONES ZOOGEOGRAFICAS Y BIOECOLOGICAS En la Fig. 15 se indica el area de distribucion conocida para cada grupo de especies de Urophonius. Si excluimos las especies de Chile central, es claro que el grupo granulatus habita un ambiente netamente patagonico, mientras el grupo brachycentrus se asocia basicamente a los antiguos macizos serranos peri- pampasicos (en el caso de U. brachycentrus , extendiendose tambien en areas no serranas, con vegetacion tipo “espinal”). Menos definida aparece el area ocupada por el grupo exochus , aunque posiblemente el registro que disponemos de el es aim muy incomplete. Como hecho destacable, parece no existir simpatria de especies pertenecientes a un mismo grupo, algo que por lo visto si seria posible entre grupos diferentes (U. jheringii y U. mahuidensis en las sierras de Tandil y Ventana; U. granulatus y U. eugenicus en Santa Cruz; U. tregualemuensis y U. sp. en Chile central). La coexistencia de U. brachycentrus y U. achalensis en las sierras de Cordoba no seria la excepcion, pues en este caso se verifica una separacion altitudinal, con distribucion de tipo parapatrico (Acosta, en prensa). De igual modo, U. brachycentrus se extiende hasta proximidades de los sistemas serranos bonaerenses -habitados por U. jheringii - pero al parecer no penetra en ellos. Otro aspecto que merece destacarse es la existencia en Urophonius de dos tipos de ciclos de actividad (Maury, 1979): algunas especies, tales como U. brachycentrus y U. jheringii , presentan un periodo de actividad “de superficie” fundamentalmente invernal, en tanto otras hacen su aparicion en los meses de verano (por ejemplo U. granulatus). Maury (1979) ha senalado que el ciclo con ACOSTA— TAXONOMIA DE UROPHONIUS 31 Fig. 15. — Distribution geografica conocida de los tres grupos de especies de Urophonius. 32 THE JOURNAL OF ARACHNOLOGY actividad invernal -invertido respect© de lo comun en otros Bothriuridae- puede interpretarse como una adaptacion secundaria, que permitiria eludir la competencia con otros escorpiones presentes en la zona (por ejemplo, las diversas especies de Bothriurus). En cuanto al grupo “estival”, el limite norte de su area de dispersion pasaria, segun Maury (1979), proximo al paralelo 46° S, que lo confinaria practicamente a la provincia argentina de Santa Cruz; la unica exception seria Urophonius sp. de la Meseta de Somuncura, ubicada casi a 500 km al norte aunque con un ambiente similar. En rigor, el periodo de actividad estival, mejor que relacionado a un area geografica dada, parece ser una caracteristica de todas las especies del grupo granulatus , ya que a U. granulatus y U. sp. de Somuncura se suma U. tregualemuensis , este ultimo con capturas entre octubre y marzo (de acuerdo con el material estudiado). En contrapartida, el periodo invernal corresponderia a las especies del grupo brachycentrus , lo que incluye a Urophonius sp. de Chile central, cuya mayor frecuencia de capturas se ubica entre abril y octubre. En el caso del grupo exochus , la information disponible es aim escasa, lo que impide por el momento tener una idea clara sobre las caracteristicas de su ciclo. AGRADECIMIENTOS Este trabajo forma parte de mi Tesis Doctoral (Facultad de Ciencias Exactas, Fisicas y Naturales, Universidad Nacional de Cordoba), dirigida por el Dr. Emilio A. Maury (Museo Argentino de Ciencias Naturales, Buenos Aires), a quien agradezco sus oportunas indicaciones, asi como sus comentarios sobre el manuscrito. LITERATURA CITADA Abalos, J. W. y C. Hominal. 1974. Urophonius achalensis, nueva especie de Bothriuridae. Acta Zool. Lilloana, 3 1(3): 19-26. Acosta, L. E. En prensa. El genero Sphaleropachylus Mello-Leitao, 1926 (Opiliones, Gonyleptidae, Pachylinae). Physis, Sec. C. Cekalovic, T. 1981. Dos nuevas especies y un nuevo registro del genero Urophonius para Chile (Scorpiones, Bothriuridae). Bol. Soc. Biol. Concepcion, 52:195-201. Francke, O. F. 1977. Scorpions of the genus Diplocentrus from Oaxaca, Mexico (Scorpionida, Diplocentridae). J. Arachnol., 4(3): 145-200. Maury, E. A. 1973. Los escorpiones de los sistemas serranos de la provincia de Buenos Aires. Physis, Sec. C, 32(85):35 1 -37 1. Maury, E. A. 1977. Comentarios sobre dos especies de escorpiones del genero Urophonius (Bothriuridae). Rev. Mus. Argentino Cienc. Nat., Entomol., 5(7): 143-160. Maury, E. A. 1979. Escorpiofauna patagonica. II. Urophonius granulatus Pocock 1898 (Bothriuridae). Physis, Sec. C, 38(94):57-68. Maury, E. A. 1980. Usefulness of the hemispermatophore in the systematics of the scorpion family Bothriuridae. Proc. 8th Int. Arachnol. Congr., Wien, 335-339. Mello-Leitao, C. 1931. Notas sobre os Bothriuridas sul-americanos. Arch. Mus. Nac. Rio de Janeiro, 33:75-113. Mello-Leitao, C. 1932. Notas sobre escorpioes sul-americanos. Arch. Mus. Nac. Rio de Janeiro, 34:9- 46. Mello-Leitao, C. 1934. Novo escorpiao brasileiro do genero Urophonius Pocock. Ann. Acad. Brasileira Sci., 6(1): 13-15. Penther, A. 1913. Beitrage zur Kenntnis Amerikanischer Skorpione. Ann. Naturh. Hofmus. Wien, 27(3):239-252. ACOSTA— TAXONOMIA DE UROPHONIUS 33 Pocock, R. 1893. A contribution to the study of Neotropical scorpions. Ann. Mag. Natur. Hist., 6th ser., 12(68):77-103. Pocock, R. 1898. Descriptions of some new scorpions from Central and South America. Ann. Mag. Natur. Hist., 7th ser., l(5):384-394. San Martin, P. 1963. Una nueva especie de Bothriurus (Scorpiones, Bothriuridae) del Uruguay. Bull. Mus. Natn. Hist. Nat., Paris, 2e. ser., 35(4):400-418. San Martin, P. 1965. Escorpiofauna argentina L Bothriuridae. Redescripcion del holotipo y description del alotipo faembra de Urophonius eugenicus (Mello-Leitao, 1931). Physis, 25(70):283- 290. San Martin, P. y T. Cekalovic. 1968. Escorpiofauna chilena I. Bothriuridae. Una nueva especie de Urophonius para Chile. Inv. Zooi. Chilenas, 13:81-100. Thor ell, T. 1877. Etudes scorpiologiques. Atti. Soc. Italian a Sci. natur., Milano, 19:75-272. Vachoe, M. 1973. Etude des caracteres utilises pour classer les families et les genres de Scorpions (Arachnides). 1. La trichobothriotaxie en Arachnologie. Sigles trichobothriaux et types de trichobothriotaxie chez les Scorpions. Bull. Mus. Nate. Hist. Nat., Paris, 3 e. ser., 140, Zool, 104:857-958. Manuscript received March 1987 , revised June 1987, Seibt, U. and W. Wickler. 1988. Interspecific tolerance in social Stegodyphus spiders (Eresidae, Araneae). J. Arachnol., 16:35-39. INTERSPECIFIC TOLERANCE IN SOCIAL STEGODYPHUS SPIDERS (ERESIDAE, ARANEAE) U. Seibt and W. Wickler Max-Planck-Institut fur Verhaltensphysiologie D-8130 Seewiesen/Starnberg, Federal Republic of Germany ABSTRACT The African social eresid spiders Stegodyphus mimosarum and S. dumicola exhibit extreme intra- as well as interspecific social tolerance. S. mimosarum individuals transferred over more than 20 km were accepted and immediately cooperated in strange conspecific colonies. In a laboratory experiment, adult females of both species formed mixed-species groups that spun and fed together. INTRODUCTION Among higher invertebrates social life has evolved in two taxa, in spiders and in insects. In spiders, social cooperation has arisen independently in several phylogenetic groups. The published schemes for the evolution of arachnid sociality suggest that two major forces may operate: a mutualistic cooperation among related or unrelated adults, and a prolongation of bonds between siblings (Buskirk 1981). The only arboreal genus in the cribellate family Eresidae, the Indo- African spider genus Stegodyphus , contains solitary as well as periodically and permanently social species, suggesting a pathway for evolution of sociality within this genus (Giltay 1927; Kullman 1972). Here we report on the social tolerance of two permanently social species, Stegodyphus mimosarum Pavesi and S. dumicola Pocock from Africa. Both species inhabit dry thornbush country, living in colonies in compact, sponge-like silk nests found mostly in thorny trees. The animals mostly rest during daylight hours. The distribution of colonies is very patchy with several km between patches. New colonies are founded by groups of nearly, or fully, adult individuals emigrating from one colony to adjacent branches and trees. In addition, single adult females “balloon” by air, presumably founding new colonies far away (Wickler and Seibt 1986). MATERIALS AND METHODS We observed the animals in the field and the laboratory. Colonies were collected in 1985 from Swaziland, Transvaal and Natal (South Africa). They can easily be kept indoors for about a year. We fed them flies, small crickets and flour beetles. 36 THE JOURNAL OF ARACHNOLOGY Tolerance tests, — To test intraspecific tolerance in the field we, on three occasions, introduced individually marked S. mimosarum females into foreign colonies more than 20 km away. In addition, we removed several individuals from different colonies and combined them into new groups in the laboratory. Interspecific tolerance was studied in- the laboratory. In order to avoid any bias from prior residence, we put five S. mimosarum and five S, dumicola adult females of similar size (6-7 mm in length; within a species, from the same colony) in an empty 10 X 10 X 10 cm glass cube without any of their original nest material Three such groups were started in parallel and observed for 55 days. We took 25 records of the spiders’ local position and social aggregations for each of the three groups (never more than one per day). Records were taken at random day-times, the spiders were always quiescent and without food at that time. RESULTS Interspecific tolerance, — In neither case did we detect differences between the contacts with strange Individuals and those between colony mates. One individual introduced into a foreign colony even joined some local individuals in subduing a prey insect within 5 min. There was no indication of colony membership identification. This result was the same as obtained in earlier experiments with the same species in Tanzania (Wickler 1973). Interspecific tolerance. — Invariably, all 10 spiders (of both species) freshly introduced into a cage formed a dense clump within 1-3 hours and remained clumped for many hours. They started spinning within one hour and the combined effort produced a silken mass. When given food, members of both species joined to subdue and consume the prey. We did not observe interspecific aggression or avoidance. In fact, all feeding groups observed were heterospecific. These groupings on food were clearly induced by the feeding situation. Since each single spider might have been attracted by the food rather than by the other spiders, these feeding groups were eliminated from the following analysis which Is based on 163 records of quiescent spider groupings. The animals were offered food about once a week; their immediate responses showed that they were hungry and, therefore, not tolerant just by satiation. Table 1 shows the frequencies of homo- and heterospecific groupings that occurred during the experiment. All 10 individuals in a cage were clumped In 27 (= 31,4%) of the 86 heterospecific groupings, forming a dense ball with maximal bodily contact. This illustrates the strong thigmotactic tendency of these spiders. Although isolated spiders of either species would attempt maximal bodily contact with any substrate (thus coming to rest In corners, fissures of bark, etc.), other Stegodyphus individuals regardless of species are more attractive. This Is an expression, of the “interattraction” typical for social spiders (Darchen 1965), Single spiders resting isolated from the other cagemates were recorded 58 times; in 45 cases it was a X dumicola , in 13 cases a S. mimosarum. The difference Is significant ai p < 0.01 (binomial test) and may have resulted from S. dumicola's higher locomotory activity. Six or more individuals were found in 65 aggregations. These necessarily contained both species. In addition, 21 groups of less than six individuals contained members of both species (Table 1). Thus, heterospecific groups were SEIBT AND WICKLER— TOLERANCE IN SOCIAL STEGODYPHUS 37 Table 1. — Frequencies of observed homo- and heterospecific groupings of Stegodyphus dumicola and S. mimosarum during 55 days. Homospecific Group size S. dumicola S. mimosarum Heterospecific 1 45 13 2 3 8 4 3 4 1 9 4 0 1 4 5 1 1 4 6 10 7 9 8 4 9 15 10 27 not a mere side effect caused by a tendency to congregate in larger groups (of more than five individuals). Groups of two to five individuals were heterospecific in 21 and homospecific in 19 cases (8 of S. dumicola , 11 of S. mimosarum ), showing no apparent tendency of either species to aggregate separately. The presence of Stegodyphus silk seems to attract individuals of either species. Searching individuals that come across a silk strand will follow it; texture and/or pheromones may be relevant cues. But two individuals, again regardless of species, coming from different directions on a completely clean surface, will contact each other in the typical manner without even touching the other’s security thread. DISCUSSION In the field we observed intermigration between separate (presumably daughter-) colonies of both species over distances less than 10 m. Bradoo (1972) reports the same phenomenon for S. sarasinorum Karsch from India. To exclude familiarity between closely neighboring groups, we mixed individuals from far distant colonies. In all cases foreign individuals were tolerated in any conspecific colony. Kullmann (1968) and Bradoo (1980) obtained the same results for S. sarasinorum . Thus there seems to be no colony integrity in social Stegodyphus spiders. Interspecific inter-colony tolerance has also been reported in the social spiders Agelena consociata Denis and A. republicana Darchen (Agelenidae), Metabus gravidus Cambridge (Araneidae), Anelosimus eximus Simon and A. studiosus Hentz (Theridiidae) and in Mallos gregalis Simon (Dictynidae) (Buskirk 1981), that is in all social species that have been so tested. Social spiders seem to differ from other social living animals in that they form open societies, in the sense that conspecific individuals are freely exchangeable between colonies. All authors theorizing on sociality in spiders (and other animals, except mixed species bird flocks and fish schools) have understood ‘social’ as something restricted to conspecifics (Wilson 1971; Vehrencamp 1979; Buskirk 1981). Social Stegodyphus spiders are believed to recognize conspecifics (Bradoo 1980). However, Kullmann et al (1971, 1972) mixed newly hatched young of the permanently social S. sarasinorum with those of the periodically or “condition- ally” (Millot and Bourgin 1942) social S. Uneatus Latreille and kept this mixed 38 THE JOURNAL OF ARACHNOLOGY group for 3.5 months. This result is supported by the observation that young individuals of even solitary spiders allow contact with members of different species (Blanke 1972). The reactions of adult individuals therefore seemed more meaningful to investigate species recognition. As the present study further shows, Stegodyphus mimosarum and S. dumicola colonies would be open even to members of the other species. The high degree of heterospecific groupings in the experimental situation indicates a considerable interspecific tolerance. Similarly, Krafft (1970, 1971) mixed the two social species Agelena consociata and A. repuhlicana (for five days under observation) which suggests that species recognition might not be relevant in this situation. He did not mention the age class of his test animals, but all age classes co-occur in Agelena colonies, so interspecific tolerance may be present in adults. Solitary spiders often live peacefully together as spiderlings and become cannibalistic later in their ontogeny. Neotenic retention of juvenile tolerance has therefore been assumed to be the first step toward communal behavior (Kullmann 1968; Buskirk 1981); it would not, however, account for interspecific tolerance. An interattraction of individuals could account for tolerance up to the point where competition would be counterselective. Under competition, selection (including kin-selection) can be expected to exclude xenogenetic individuals from tolerance. However, an individual’s decision to attack or tolerate a stranger would still be governed by a cost/ benefit ratio. For a socially living individual the cost factor may be most important: attacking will provoke defensive counteraggres- sion, and the full risk of being severely damaged would fall upon the attacking individual, while costs arising from tolerance would be shared among all community members. Mixed species Stegodyphus colonies are unknown from the field, perhaps because no one has looked for them. Both species co-occur closely in Transvaal, and the nearest interspecific colony distance that we encountered was 5 m within the same tree. On the other hand, our observations of the spiders suggest that the two Stegodyphus species would eventually separate according to their different behaviors (including walking speed, reaction times, etc.). S. mimosarum tends to live higher up in trees, while S. dumicola colonies are typically found closer to the ground (Seibt and Wickler 1988). Similarly in the genus Agelena , A. consociata prefers shadowy zones between lower bushes, while A. repuhlicana builds its colonies in the crowns of trees exposed to the sun (Krafft 1970, 1971). Thus in both cases an ecological separation seems to counteract heterospecific groupings. ACKNOWLEDGMENTS We thank the National Parks Board of Trustees in Pretoria as well as the Natal Parks, Game and Fish Preservation Board for permission to work in the Nature Reserves. The comments of R. Buskirk, W. B. Peck and W. J. Tietjen in their review of the manuscript were greatly appreciated. SEIBT AND WICKLER — TOLERANCE IN SOCIAL STEGODYPHUS 39 LITERATURE CITED Blanke, R. 1972. Untersuchungen zur Okophysiologie und Okoethologie von Cyrtophora citricola Forskal (Araneae, Araneidae) in Andalusien. Forma et Functio, 5:125-206. Bradoo, B. L. 1972. Some observations on the ecology of social spider Stegodyphus sarasinorum Karsch (Araneae: Eresidae) from India. Oriental Insects, 6:193-204. Bradoo, B. L. 1980. Feeding behaviour and recruitment display in the social spider Stegodyphus sarasinorum Karsch (Araneae, Eresidae). Tijdschr. Entomol., 123:89-104. Buskirk, R. E. 1981. Sociality in the Arachnida. Pp. 281-367, In Social Insects (H. R. Hermann, ed.). Vol II, Academic Press, New York. Darchen, R. 1965. Ethologie de quelques araignees sociales: l’interattraction, la construction et la chasse. C. R. 5. Congr. intern pour Petude des insectes sociaux, Toulouse, pp. 333-345. Giltay, L. 1927. Une Araignee sociale du Kasai ( Stegodyphus simoni nov. sp.) Rev. Zool. Afrique, 15:105-117. Krafft, B. 1970, 1971. Contribution a la biologie et a Fethologie d 'Agelena consociata Denis (Araignee sociale du Gabon), II, III. Biologia Gabonica, 6:199-301, 307-369; 7:3-56. Kullmann, E. 1968. Soziale Phaenomene bei Spinnen. Insectes Sociaux, 15:289-298. Kullmann, E. 1972. Evolution of social behavior in spiders (Araneae, Eresidae and Theridiidae). Amer. Zool., 12:419-426. Millot, J. et P. Bourgin. 1942. Sur la biologie des Stegodyphus solitaires (Araneides, Eresides). Bull. Biol. France et Belgique, 76:299-314. Seibt, U. and W. Wickler. 1988. Bionomics and social structure of ‘Family Spiders' of the genus Stegodyphus, with special reference to the African species S. dumicola and S. mimosarum (Araeeida, Eresidae). Verb, naturwiss. Ver. Hamburg (N.F.), 30 (in press). Vehrencamp, S. L. 1979. The roles of individual, kin and group selection in the evolution of sociality. Pp. 351-394, In Social Behavior and Communication. (P. Marler, J. G. Vandenbergh, eds.). Handbook of Behavioural Neurobiology III, Plenum Press, New York/ London. Wickler, W. 1973. Uber Koloniegriindung und soziale Bindung von Stegodyphus mimosarum Pavesi und anderen sozialen Spinnen. Z. Tierpsychol., 32:522-531. Wickler, W. and U. Seibt. 1986. Aerial dispersal by ballooning in adult Stegodyphus mimosarum. Naturwissenschaften, 73:628-629. Wilson, E. O. 1971. The Insect Societies. Harvard Univ. Press, Cambridge Massachusetts. Manuscript received March 1987, revised June 1987. - Downes, M. F. 1988. The effect of temperature on oviposition interval and early development in Theridion rufipes Lucas (Araneae, Theridiidae). J. Arachnoh, 16:41-45. THE EFFECT OF TEMPERATURE ON OVIPOSITION INTERVAL AND EARLY DEVELOPMENT IN THERIDION RUFIPES LUCAS (ARANEAE, THERIDIIDAE) Michael F. Downes Zoology Department James Cook University of North Queensland Townsville, Qld. 4811, Australia ABSTRACT The effect of temperature on oviposition rate and early development of Theridion rufipes in northern Queensland was investigated. Development did not proceed at 20° C, and embryos and postembryos responded differently to temperatures of 25° C and 30° C. The time interval between ovipositions by the female was markedly extended at 20° C, a temperature at which development was unlikely. INTRODUCTION Spider species for which the relationship between temperature and development has been reported include Tegenaria atrica (C. Koch) (Browning 1941), Cheiracanthium inclusum (Hentz) (Peck and Whitcomb 1970), Thanatus striatus (C. Koch) and Allomengea scopigera (Grube) (Schaefer 1977) and Latrodectus hasselti Thorell (Downes 1987). The present study adds Theridion rufipes Lucas to this list, and shows that the oviposition intervals (i.e., the time periods between the construction of egg sacs, over the iteroparous sequence) are temperature- dependent in a comparable way. T. rufipes is widely distributed, primarily in tropical and subtropical regions, but with a range extending into temperate zones; it was first described from specimens collected in Algeria. Its original range may be hard to determine because its close association with man has probably led to extensive transportation. A recent first report (Sugarman 1979) of its occurrence in the Marshall Islands, for instance, may reflect its previous introduction by man. In the United States its range includes Texas and Florida (Levi and Randolph 1975). MATERIALS AND METHODS A total of 60 T. rufipes females was collected in Townsville, of which 51 were immature and unmated and nine were mature and had mated and produced their first egg sac in the field. These females were separated into three groups of 20 (17 immature, three mature in each case), and each group was kept at one of three experimental temperatures (20, 25 and 30° C ± 1°C). The photoperiod was in 42 THE JOURNAL OF ARACHNOLOGY each case 14/10 hours light/ dark. After their maturation molt, the previously immature females were mated, and on those (31) occasions when mating took place directly, and the male was either then killed by the female or removed, it was possible to record the time from mating to first oviposition. Times between subsequent ovipositions of all females (lab-mated and field-mated) were routinely recorded. Spiders were confined individually in glass tubes measuring 50 X 20 mm diameter, with perforated plastic stoppers. Twice weekly, each was fed an identical diet of insect prey, including Drosophila sp., muscoid flies and native cockroaches. Water was not provided. Observations were normally made daily; when ovipositions, hatchings, molts or emergences coincided with missed daily inspections, these data were not included in the results. Hence, although the spiders produced 338 egg sacs, only 287 of these gave oviposition interval values and 249 provided development data at 25 and 30° C; those (45) sacs that were constructed at 20° C gave the negative results for development at that temperature. All sacs were incubated at the same temperatures at which they were constructed. Embryonic development times were obtained from 60 egg sacs which were teased open and housed in glass cavity blocks, the glass covers of which were separated from the rims of the blocks by a layer of non-absorbent cotton wool, with a little vaseline as adherent. By pseudoreplication, 40 of these same sacs gave postembryonic development times. The time from first molt to emergence was obtained from 189 egg sacs which were transferred to fresh 50 X 20 mm glass tubes with perforated plastic stoppers (these containers were also used for the 20° C sacs mentioned above), and kept intact until emergence occurred. The interval between the first molt and emergence was calculated by subtraction of the oviposition-first molt time (as determined from teased-open sacs) from the oviposition-emergence time of the intact sacs. Some possible inaccuracies of this procedure were considered by Downes (1987). All developmental data are for whole broods (i.e., egg sacs) rather than for individual spiderlings. Although individuals varied in their rate of development, developmental synchrony was very close at the temperatures at which development occurred in this study. The values for the interval between oviposition and emergence of spiderlings from the egg sac were necessarily whole- brood values, so it was felt justifiable to use whole-brood means to compute corresponding values for intervals between oviposition and hatching (= embryo) and between hatching and the first molt (= postembryo). Hatch time was defined as the time of the rupture of the chorion. Field temperatures were provided by the Geography Department of James Cook University. RESULTS Development did not proceed at 20° C. The mean duration, at 25 and 30° C, of the embryonic, postembryonic and first instar stages within the egg sac (i.e., from oviposition to hatching, molting and emergence respectively) are given in Table L The temperature increase of 5°C from 25° C produced a decrease in development time of 2.7 days (28%) for the embryonic stage but only 0.3 days (14%) for the postembryonic stage. DOWNES— OVIPOSITION AND DEVELOPMENT IN THERIDION 43 Table 1. — The early development of Theridion rufipes. Values are given as mean interval in days, with standard error and (sample size). Cumulative values (cumul.) are given as values only. Values are means of whole broods (i.e., egg sacs), not of individual spiderlings. Duration of Developmental Stages Developmental stage 20° C 25° C cumul. 30° C cumul. Embryo No devel. (45) 9.8 SE 0.18 (45) 9.8 7.1 SE 0.27 (15) 7.1 Postembryo — 2.1 SE 0.22 (27) 11.9 1.8 SE 0.37 (13) 8.9 First instar to emergence — 2.4 SE 0.07 (124) 14.3 3.2 SE 0.27 (65) 12.1 Despite the fact that the overall development time from oviposition to emergence decreased by 2.2 days (15%) with the same temperature change (Table 1, cumulative values), the within-sac first instar stage time actually increased by 0.8 days (33%), from 2.4 to 3.2 days (Table 1, non-cumulative values). There was no evidence that the adult female spiders were adversely affected by the experimental temperature extremes of 20 and 30° C, except that matings were rarer at the former temperature. However, the oviposition sequence was significantly extended at 20° C. Times, in days, separating the ovipositions of the iteroparous sequence are presented in Table 2. It is unfortunate that so few (three) instances were recorded of mating-oviposition at 20° C because the mean of these few values is by far the highest of those presented. The time between mating and the first oviposition is much shorter than that between subsequent ovipositions at temperatures that favor normal early development, but the data of Table 2 suggest that the reverse may be true at a temperature at which development is not assured. DISCUSSION As in all poikilotherms, temperature is usually the most critical environmental factor influencing the rate of development. However, an animal rarely develops under a constant unchanging temperature regime but rather experiences daily temperature variation around a gradually changing seasonal mean. Latrodectus hasselti , for example, commonly experiences temperatures of 30° C or greater in Townsville, but 30° C is close to an upper limit of temperature tolerance if applied continuously (Downes 1987). The relative duration of the three pre-emergence phases of F rufipes (embryo, postembryo and the first instar (part)) alters with the temperature change. The embryonic period takes up 69% of the total time to emergence at 25° C but only 59% of the total time at 30° C. The relative duration of the much shorter Table 2. — The effect of temperature on time from mating to first oviposition and between subsequent ovipositions in Theridion rufipes. Values are given as mean interval in days, with standard error and (sample size). Temperature (°C) Mating to first oviposition Subsequent oviposition intervals 20 47.7 SE 0.37 (3) 26.2 SE 2.16 (38) 25 6.6 SE 1.03 (17) 16.1 SE 0.57 (123) 30 6.3 SE 1.08 (11) 14.1 SE 0.61 (95) 44 THE JOURNAL OF ARACHNOLOGY postembryonic period remains unaltered (15% of the total time), while the relative (and indeed the absolute) time from the first ecdysis to emergence increases from 17% at 25° C to 26% at 30° C. This does not in itself imply any difference in the physiological response to temperature between embryos and postembryos, or between embryos and first instar spiderlings; it may be that the first instar spiderlings respond to high temperatures by remaining in the egg sac. This would be advantageous because the spiderlings would desiccate more rapidly once out of the cocoon owing to their increased exposure and increased activity. If this is so, moderately high temperatures may prolong the time spent in the egg sac by the first instars; this would explain the apparent increase in development time of the first instars in Table 1 . In fact this phase is not a developmental stage in the same sense as the two earlier stages; the period from first to second ecdysis, rather than that from first ecdysis to emergence, would be more comparable. It is not clear why the embryo and postembryo stages respond differently to a temperature decrease of 5°C from. 30° C, the embryo stage increasing its development time by 38% (2.7 days) and the postembryo stage by only 17% (0.3 days). These differences may relate to the (unknown) overwintering strategy of this species in temperate zones. A ‘stronger’ response from embryos than from postembryos suggests physiological affinities with populations that undergo egg overwintering, but most spider species do not overwinter in the egg stage, according to Foelix (1982). In temperate regions T rufipes may be found to develop (slower) at temperatures down to 10° C or below, as does Latrodectus hasselti from temperate zones (Forster 1984). Why this does not occur in tropical populations is an open question, but over the past six years the mean monthly temperature in this district of Townsville has fallen below 19°C only in one two-month period, when it sank to 18.3°C (June and July 1982). Minimum temperatures do occasionally go below 10° C during some nights of the cooler dry season, but corresponding day temperatures on these occasions are likely to be in excess of 20° C. A temperature cycle of 1 5-25° C, evenly changing over a 24-hour period, may produce a very different outcome from a constant 20° C regime, despite the former’s mean value of 20° C. In view of the curtailment of development at 20° C, the data of Table 2 suggest that there may be a behavioral response on the part of T rufipes females, delaying the production of egg sacs at temperatures at which development is unlikely. In a poikilotherm such as T, rufipes physiological processes such as oogenesis will be subject to the effects of temperature just as surely as will embryonic development, but it would be interesting if it could be demonstrated that ovipositioe behavior responds more strongly than physiological/ metabolic processes, to a fall in temperature of 5°C from 25 to 20° C. Such an investigation was beyond the scope of the present study. ACKNOWLEDGEMENTS I thank Professor Rhondda Jones who supervised the M.Sc. research project of which this paper represents a part. Helpful comments from Dr. Robert Jackson and Dr. James E. Carrel have enabled me to improve the manuscript. Dr. Valerie Davies identified the spider. DOWNES— OVIPOSITION AND DEVELOPMENT IN THERIDION 45 LITERATURE CITED Browning, H. C. 1941. The relation of instar length to the external and internal environment in Tegenaria atrica (Arachnida). Proc. Zool. Soc. London, Ser. A, 111:303-317. Downes, M. F. 1987. Postembryonic development of Latrodectus hasselti Thorell (Araneae, Theridiidae). J. Arachnol., 14:293-301. Foelix, R. F. 1982. Biology of Spiders. Harvard Univ. Press, Cambridge, Mass., 306 pp. Forster, L. M. 1984. The Australian redback spider ( Latrodectus hasselti ): its introduction and potential for establishment and distribution in New Zealand. Pp. 273-289, In Commerce and the Spread of Pests and Disease Vectors. (M. Laird, ed.). Praeger, New York. Levi, H. W. and D. Randolph. 1975. A key and checklist of American spiders of the family Theridiidae north of Mexico (Araneae). J. Arachnol., 3:31-51. Peck, W. B. and W. H. Whitcomb. 1970. Studies on the biology of a spider, Chiracanthium inclusum (Hentz). Univ. Arkansas Agric. Exp. Sta. Bull., 753:1-76. Schaefer, M. 1977. Untersuchungen uber das wachstum von zwei spinnenarten (Araneida) im labor und frieland. Pedobiologia, 17:189-200. Sugarman, B. B. 1979. Additions to the list of insects and other arthropods from Kwajalein Atoll (Marshall Islands). Proc. Hawaii Entomol. Soc., 13:147-151. Manuscript received March 1987, no revision. » Beatty, J. A. and J. W. Berry. 1988. The spider genus Paratheuma Bryant (Araneae, Desidae). J. Arachnol., 16:47-54. THE SPIDER GENUS PARATHEUMA BRYANT (ARANEAE, DESIDAE) Joseph A. Beatty Department of Zoology Southern Illinois University Carbondale, Illinois 62901-6501 USA and James W. Berry Department of Biological Sciences Butler University Indianapolis, Indiana 46208 USA ABSTRACT The genus Swainsia Marples, 1964, is newly synonymized with Paratheuma Bryant, 1940. The previously unknown male of P insulana (Banks) and female of P. armata (Marples) are described and illustrated. Habitat data and new distribution records are provided for these two species. The family Desidae is reported from the United States for the first time. INTRODUCTION Members of the genus Paratheuma are rare and poorly known spiders. The primary literature on them is limited to six papers (Banks 1902, 1903; Bryant 1940; Marples 1964; Roth and Brown 1975; Platnick 1977); the total number of specimens reported is only 19, although Roth and Brown (1975) apparently did not list all the specimens they collected, and Banks (1903) gave no data on number of specimens. This apparent rarity is no doubt a result of the restricted littoral and intertidal habitat of the spiders. Where found, they may be reasonably common, as indicated by the records presented below. Only two species have hitherto been recognized as belonging to the genus (Platnick 1977): Eutichurus insulanus Banks, 1902, from Cuba, Haiti and Bermuda; and Corteza inter aesta Roth and Brown, 1975, from the Gulf of California. To these may now be added a third species, Swainsia armata Marples, 1964, from Swains Island, American Samoa. It is remarkable that these three congeneric species have been described not only in separate genera, but in separate families — Eutichurus insulanus in the Clubionidae (later transferred to the Gnaphosidae), Corteza in the Desidae, and Swainsia in the Agelenidae. It is quickly evident from the species’ morphology that the only classical family in which they could be placed is the family Agelenidae. Their assignment to the Clubionidae and Gnaphosidae can be 48 THE JOURNAL OF ARACHNOLOGY regarded only as a blunder, even by the standards of the time, resulting from careless observation. The family Desidae, in which the genus is now placed, has received widespread acceptance only recently. Bryant (1940) established the genus Paratheuma for E. insulanus and P. isolata, incorrectly placing the genus in the Gnaphosidae, and including erroneous observations in the generic description. Paratheuma isolata has been transferred to the genus Syrisca (Clubionidae). Genus Paratheuma Bryant Paratheuma Bryant 1940:387, type species by original designation Eutichurus insulanus Banks. Roewer 1954:353. Platnick 1977:199. Swainsia Marples 1964:403, type species by monotypy Swainsia armata Marples. Brignoli 1983:52. NEW SYNONYMY. Corteza Roth and Brown 1975:2, type species by original designation and monotypy Corteza interaesta Roth and Brown. First synonymized with Paratheuma by Platnick 1977:199. Description. — Color in life. Chelicerae orange brown to dark brown. Endites, labium and cephalic portion of carapace orange brown to brown, thoracic portion of carapace lighter. Sternum yellow orange to orange brown. Legs yellow to gray. Abdomen grayish yellow to greenish gray, often with small yellowish flecks or indistinct chevrons flanking cardiac region in posterior 2/3 of dorsum. Carapace may have dusky patches or lines along intercoxal grooves. In alcohol color may fade so that abdomen and legs are nearly uniform pale yellow, carapace yellow brown. Total length of females 3. 2-5. 9 mm, males 3. 1-5.2 mm. Carapace length females 1.35-2.50 mm, males 1.50-2.72 mm. Carapace width (maximum) females 1.00-2.00 mm, males 1.05-2.04 mm. Carapace low, cephalic region not much higher than thoracic, and sloping gradually posteriad. Cephalic region width about 2/3 of maximum carapace width or slightly less. Eye rows virtually straight, or with anterior row slightly procurved. Eyes subequal in size, AME slightly smaller than others. Width of posterior eye row about 70% of head width, anterior row slightly narrower. AME dark, others light. Chelicerae stout, somewhat geniculate at base, divergent distally. Male chelicerae somewhat porrect and more divergent than in females. Chelicerai surface mostly clothed with abundant short setae (shorter in males) with setal bases somewhat enlarged. Promarginal chelicerai teeth in proximal half of fang groove only, two large teeth and one smaller. Retromarginal teeth extending full length of groove, 5-8 small teeth, largest tooth distal. Endites rectangular, about 1.5 times as long as wide, bluntly pointed distally, with conspicuous scopula of medially curving hairs. Labium short and broad, rounded to slightly emarginate distally. Sternum slightly longer than broad to as broad as long, truncate, extending to posterior edge of coxae 4, narrowed and rounded posteriorly. Hind coxae separated by width of one coxa or slightly less. Legs slender. Femoral lengths of all legs shorter than to as long as carapace length in females, shorter to slightly longer in males. Leg length formula 4-1-2-3. Leg spines few, weak. Male palp slender, cymbium only slightly wider than distal width of tibia. Cymbium narrowed, finger-like distally, bulb occupying less than to more then half its length. Palp relatively simple, embolus and conductor chief parts visible in ventral view. Embolus slender, originating near middle of medial edge of bulb, or BEATTY AND BERRY — PARATHEJJMA 49 somewhat more basally, curving forward to alveolar margin and back to form almost complete circle, ending on pointed conductor, which extends proximally to overlap tibia. Tibia with short to long disto-medial apophysis, and sometimes short distolateral apophysis as well. Abdomen oval, unmodified, thickly covered with short, fine, appressed hairs. Posterior spinnerets elongate, up to half length of abdomen, extending beyond abdominal tip 0.5- 1.0 mm; 2-segmented, segments about equal in length. Anterior spinnerets 2-segmented, distal segment very short, separated basally by about width of spinneret. Median spinnerets shorter, 1 -segmented. Colulus large, conspicuous, heavily setose (colulus mistaken for a “lobe of spiracle” by Bryant, 1940, who described the genus as lacking a colulus). Tracheal spiracle broad, occupying about as much of abdominal width as pair of anterior spinnerets, separated from colulus by distance about equal to length of colulus. Epigynum consisting of two small to large depressions, long axes varying from transverse to nearly longitudinal, with either edges only or entire surfaces sclerotized. Openings obvious or indistinct, ducts and receptacles appearing as two short to long longitudinal dark areas near medial edges of the depressions. Internal structures relatively simple (Fig. 6, see also Figs. 2, 4 in Platnick 1977). Overall epigynal size smallest in P insulana , largest in P. armata. Diagnosis. — The extension of the tracheae into the thorax, the absence of tarsal scopulae, the presence of a single row of tarsal trichobothria, and the genital structure separate the genus from other littoral genera of the family. Paratheuma insulana (Banks) Figs. 1,4,7, 10 Eutichurus insulanus Banks 1902:270 (female holotype from the Bermuda Islands, lost). Bonnet 1956:1845. Paratheuma insulana (Banks), Bryant 1940;387; Roewer 1954:353; Platnick 1977:200. Diagnosis. — The small size and oblique orientation of the epigynal depressions of the female, and the bifurcate, laterally projecting conductor tip of the male distinguish P. insulana from the other members of the genus. Male. — Total length 3. 5-5.0 mm, mean 3.86, SE 0.130. Carapace length 1.65- 2.10 mm, mean 1.793, SE 0.039. Carapace width 1.15-1.55 mm, mean 1.290, SE 0.034 (ten specimens measured). Other measurements of one male: head width 0.85 mm, sternum length 1.0 mm, sternum width 1.0 mm, endite length 0.6 mm, labium length 0.35 mm, leg I — femur 1.75 mm, patella-tibia 2.25 mm, metatarsus- tarsus 2.45 mm, leg II — femur 1.75 mm, patella-tibia 2.1 mm, metatarsus-tarsus 2.4 mm, leg III — femur 1.55 mm, patella-tibia 1.9 mm, metatarsus-tarsus 2.35 mm, leg IV — femur 1.9 mm, patella-tibia 2.45 mm, metatarsus-tarsus 3.05 mm. Bulb of palp smaller than that of other species, occupying less than 1/2 length of cymbium. Conductor bifurcate distaliy, with one branch extending laterally (Fig. 1). Female. — Total length 3. 3-5. 9 mm, mean 4.32, SE 0.236. Carapace length 1.35- 2.00 mm, mean 1.730, SE 0.058. Carapace width 1.00-1.45 mm, mean 1.238, SE 0.042 (ten specimens measured). Other measurements of one female: head width 0.9 mm, sternum length 0.95 mm, sternum width 0.9 mm, endite length 0.5 mm, labium length 0.35 mm, leg I — femur 1.55 mm, patella-tibia 2.0 mm, metatarsus- 50 THE JOURNAL OF ARACHNOLOGY Figs. 1-3. — Left pedipalps of male Paratheuma in ventral view: 1, P. insulana from Pigeon Key; 2, P. interaesta from Puerto Penasco, Sonora, Mexico (holotype); 3, P. armata from Eniwetok, Marshall Islands. tarsus 2.25 mm, leg II — femur 1.45 mm, patella-tibia 1.75 mm, metatarsus-tarsus 2.10 mm, leg III — femur 1.30 mm, patella-tibia 1.65 mm, metatarsus-tarsus 2.00 mm, leg IV — femur 1.70 mm, patella-tibia 2.20 mm, metatarsus-tarsus 2.75 mm. Depressions of epigynum shorter and narrower than in other species, oriented obliquely. Internal genitalic structures are smaller than in others (Fig. 7). Distribution. — Reported from Cuba (Bryant 1940), Haiti (Banks 1903), and the Bermudas (Banks 1902). Also known from the Florida Keys, Monroe Co., Florida, USA, Key Largo to West Summerland Key. Habitat. — All specimens for which habitat data are available were collected among coral rubble (sometimes buried under plant detritus) above and below high water mark on sea beaches. Material examined. — FLORIDA; Monroe Co ., Key Largo, 30 Dec. 1984, (J. W. Berry), 1 male, 2 females; Pigeon Key, 12 Mar. 1981, (J. W. Berry), 5 males, 5 females, 13 immatures, 25-28 Dec. 1984, (J. W. Berry), 1 male, 8 females, 63 immatures, 14 Mar. 1985, (J. W. Berry), 6 males, 7 females; Big Pine Key, 18 June 1965, 1 female, 5 immatures, in tidal litter, (W. Suter-FMNH), 28-29 Dec. 1984, (J. W. Berry), 1 female, 1 immature; West Summerland Key, 28 Dec. 1984, (J. W. Berry), 2 males. Paratheuma interaesta (Roth and Brown) Figs. 2, 5, 8, 10 Corteza interaesta Roth and Brown 1975:3 (male holotype and female paratype from Pelican Point, Sonora, Mexico, in the American Museum of Natural History, examined). Paratheuma interaesta : Platnick 1977:200. Diagnosis. — The intermediate size of the genitalia of both sexes, transversely oriented main axis of the epigynal depressions of the female, and short tibial apophysis and simply pointed conductor of the male distinguish this species from the other two. No additional information is available for this species, except for the collection of an immature specimen at the type locality, by Beatty, on 28 July 1962. The male palp and female epigynum are illustrated (Figs. 2, 5, 8) for comparison with the other species. BEATTY AND BERRY — PARATHEUMA 5! Figs. 4-9. — Epigyna of female Paratheuma : 4-6, ventral views; 7-9, dorsal views, cleared; 4, 7, P. insulana from Pigeon Key; 5, 8, R interaesta from Puerto Penasco, Sonora, Mexico (paratype); 6, 9, P. armata from Eniwetok. Material examined. — MEXICO: SONORA; Norse Beach, Pelican Point, 27 March 1969, (V. Roth), 1 male, 1 female (faoiotype and paratype), 28 July 1962, (J. A. Beatty), 1 female, 1 immature. Paratheuma armata (Marples), new combination Figs. 3, 6, 9, 11 Swaimio armata Marples 1964:403 (male holotype from Swains Island, American Samoa, in Bishop Museum, Honolulu, examined). Brignoli 1983:521. Diagnosis. — The relatively large epigynum of the female, which has large round depressions with the axis of depression plus ducts nearly longitudinal, and the very long medial and short lateral apophyses of the male palp clearly separate this species from P. interaesta and P. insulana, Male. — Total length 3. 1-3.8 mm, mean 3.45, SE 0.146. Carapace length 1.50- 1.65 mm, mean 1.590, SE 0.026. Carapace width 1.05-1.25 mm, mean 1.180, SE 0.033 (five specimens measured). Other measurements of one male: head width 0.80 mm, sternum length 0.85 mm, sternum width 0,85 mm, endite length 0.50 mm, labium length 0.25 mm, leg I — femur 1.65 mm, patella-tibia 2.00 mm, metatarsus-tarsus 2.25 mm, leg II — femur 1.63 mm, patella-tibia 1.85 mm, metatarsus-tarsus 2.10 mm, leg III — femur 1.50 mm, patella-tibia 1.65 mm, metatarsus-tarsus 2.05 mm, leg IV — femur 1.75 mm, patella-tibia 2.10 mm, metatarsus-tarsus 2.60 mm. Bulb of palp (Fig. 3) larger than in other species, occupying more than half length of cymbium and, when flexed, overlapping tibia by almost half length of cymbium; short, broad, collar-like lateral apophysis is present. Female. — Total length 3. 5-5.0 mm, mean 4.08, SE 0.198. Carapace length 1 50- 1.90 mm, mean 1.658, SE 0.058. Carapace width 1.10-1.40 mm, mean 1.24, SE 0.048 (six specimens measured). Other measurements of one female: head width 1.05 mm, sternum length 1.0 mm, sternum width 0.95 mm, endite length 0.60 mm, labium length 0.35 mm, leg I — femur 1.70 mm, patella-tibia 2.15 mm, metatarsus-tarsus 2.30 mm, leg II — femur 1.65 mm, patella-tibia 2.00 mm, 52 THE JOURNAL OF ARACHNOLOGY Fig. 10. — Distribution of Paratheuma insulana (circles) and Paratheuma interaesta (triangles). metatarsus-tarsus 2.35 mm, leg 111 — femur 1.55 mm, patella-tibia 1.85 mm, metatarsus-tarsus 2.10 mm, leg IV — femur 1.90 mm, patella-tibia 2.35 mm, metatarsus-tarsus 2.75 mm. Epigynum (Fig. 6) much larger than in other species, about 1 mm broad by 2/3 mm long, occupying entire width of area between book lungs, and most of distance from base of pedicel to epigastric groove. Depressions broader, more widely separated than in other species; axis of depressions plus ducts almost parallel with midliee. Entire depression, rather than edges only, sclerotized. Internal geeitalic structures (Fig. 9) more extensive than in other species. Distribution. — From American Samoa to the Marshall Islands and western Caroline Islands (Fig. 11). Habitat. — All specimens for which data are available were taken among broken coral or other beach rubble near the upper edge of the drift line on sea beaches. Material examined. — AMERICAN SAMOA: Swains Island , 20 Aug. 1940, (E. C. Zimmerman), 1 male (holotype, BISH). MARSHALL ISLANDS: Eniwetok Atoll , 9 June 1969, (J. W. Berry), 2 immature, 16 July 1968, (J. W. Berry and J. A. Beatty), I male, 1 female, 10 immature (SIU); Kwajalein Atoll , 8 Aug. 1969, (J. W. Berry), 6 immature (SIU); Majuro Atoll , I Aug. 1969, (J. W. Berry), 3 female, 12 immature, 31 July 1969, (J. W. Berry), 1 immature (SIU). CAROLINE ISLANDS: PALAU; Kayangel Atoll, 23 May 1973, (J. W. Berry and E. Berry), 1 male, 1 female, 3 BEATTY AND BERRY— PA RATH EUM A 53 11 Fig. 11. — Distribution of Paratheuma armata. immature (SIU); Palau District , Pulo Anna Isl., 7 Apr. 1973, (J. W. Berry and E. Berry), 1 male, 1 female, 6 immature (SIU), Sonsorol Isl., 10 Apr. 1973, (J. W. Berry and E. Berry), 1 male, 4 immature; YAP, 11 May 1980, (J. W. Berry), 1 male, 16 immature (SIU), Ulithi Atoll , 2 May 1980, (J. W. Berry and E. Berry), 1 male, 1 female (SIU). DISCUSSION The habitat, general appearance, and specific structures such as the genitalia, spinnerets, colulus, and tracheal system clearly place Swainsia armata with the other members of the genus Paratheuma. The taxonomic misplacement of Paratheuma undoubtedly was the primary reason for its being overlooked by the authors of the synonymous genera. Lack of definite locality or habitat data account for the rarity of P insulana and P. armata in collections, as both are rather common to abundant in suitable habitats. Although no specific collecting locality was given for the holotype of P insulana , some information can be derived from the name of the collector, W. G. VanName, and the apparent date, May 1901. VanName was a specialist in tunicates and isopods, and collected at several specific localities in the Bermudas in May 1901. These localities are listed by him (VanName 1902) and, should it become desirable to restrict the type locality, one of them should probably be chosen. Specimens of P. insulana and P. armata will be deposited in the collections of the American Museum of Natural History, New York, New York, the Museum of 54 THE JOURNAL OF ARACHNOLOGY Comparative Zoology, Cambridge, Massachusetts, and the Bishop Museum, Honolulu, Hawaii, (BiSH) The remainder of the material will be kept in the Research Museum of the Department of Zoology, Southern Illinois University at Carbondale, Carbondale, Illinois, (SIU), and the Department of Biological Sciences, Butler University, Indianapolis, Indiana, (BU). The abbreviation FMNH was used for the Field Museum of Natural History. ACKNOWLEDGMENTS We are indebted to Dr. N. I. Platnick for loan of the type and paratype of Corteza interaesta. Elizabeth Berry provided invaluable logistic support and was of assistance in all phases of the work. This research was in part supported by an AEC grant to the University of Hawaii, for study at the Eniwetok Marine Biological Laboratory. Preparation of the manuscript was facilitated by a Butler University Fellowship awarded to the second author. LITERATURE CITED Banks, N. 1902. Some spiders and mites from the Bermuda Islands. Trans. Connecticut Acad. Arts Sci., 1 1:267-275. Banks, N., 1903. A list of Arachnida from Hayti, with descriptions of new species. Proc. Acad. Nat. Sci. Philadelphia, 55:340-345. Bonnet, P. 1956. Bibliographia Araneorum. Toulouse, 2:919-1925. Brignoli, P. M. 1983. A Catalogue of the Araneae Described Between 1940 and 1981. Manchester. 755 PP Bryant, E. B. 1940. Cuban spiders in the Museum of Comparative Zoology. Bull. Mus. Comp. ZooL. 86:249-532. Marples, B. J. 1964. Spiders from some Pacific Islands, part V. Pacific Sci., 18:399-410. Platnick, N. I. 1977. Notes on the spider genus Paratheuma Bryant (Arachnida, Araneae). J. ArachnoL, 3:199-201. Roewer, C. F. 1954. Katalog der Araneae. Bremen. 2:1-923. Roth, V. D. and W. L. Brown. 1975. A new genus of Mexican intertidal zone spider (Desidae) with biological and behavioral notes. Amer. Mus. Novitates, 2568:1-7. VanName, W. G. 1902. The Ascidians of the Bermuda Islands. Trans. Connecticut Acad. Arts Sci., 11:325-411. Manuscript received April 1987, revised June 1987 . Jennings, D. T., M. W. Houseweart, C. D. Dondale and J. H. Redner. 1988. Spiders (Araneae) associated with strip-clearcut and dense spruce-fir forest of Maine. J. Arachnol., 16:55-70. SPIDERS (ARANEAE) ASSOCIATED WITH STRIP-CLEARCUT AND DENSE SPRUCE-FIR FORESTS OF MAINE1 Daniel T. Jennings Northeastern Forest Experiment Station USDA Building, University of Maine Orono, Maine 04469 USA Mark W. Houseweart Cooperative Forestry Research Unit College of Forest Resources University of Maine Orono, Maine 04469 USA and Charles D. Dondale and James H. Redner Biosystematics Research Centre Research Branch, Agriculture Canada Ottawa, Ontario K1A 0C6 Canada ABSTRACT Spiders of 15 families, 76 genera, and at least 125 species were collected by pitfall traps in a spruce- budworm infested forest of northern Maine. Species of Lycosidae were numerically dominant and accounted for 56.2 and 54.1% of the total trapped specimens in 1977 and 1978, respectively. For both study years, significantly more (P < 0.05) individuals and species of spiders were captured in clearcut strips than in either uncut residual strips or dense stands. Peaks in seasonal activity of spiders generally coincided with the spruce budworm’s early and late larval stages; spiders were also abundant and active during budworm oviposition and dispersal of 1st instars. Diversity of spider species was generally greater in dense stands and uncut residual strips than in clearcut strips. Individuals were distributed unevenly among species but more evenly in dense stands and uncut residual strips than in clearcut strips. Coefficients of community (CC) and percentage similarity (PS) values indicated more spider species than individuals were shared in common among forest conditions. Neither age of strip clearcut (1-6 years) nor litter depth had much influence on mean catches and mean numbers of species of spiders per trap per week. INTRODUCTION Spiders are among the dominant predators in many terrestrial communities (Gertsch 1979). In northeastern spruce-fir forests, arboreal spider densities are estimated to range from 187,500/ha (Morris 1963) to 312,500/ha (Haynes and 'Mention of a commercial or proprietary product does not constitute endorsement by the U.S. Department of Agriculture, Forest Service, University of Maine, or Agriculture Canada. 56 THE JOURNAL OF ARACHNOLOGY Sisojevic 1966). These estimates do not include the epigeal and terricolous faunas that live near the ground. And, despite their common occurrence and potential importance as predators of insect pests (Riechert 1974), little is known about the species composition, diversity, and abundance of spiders that inhabit individual forest stands, forest-stand types, or forest communities in North America. Some earlier studies of forest-spider faunas include those of Dowdy (1950), Elliott (1930), Gibson (1947), Stratton et al (1979), and Uetz (1979). The possible adverse or beneficial effects of forest management practices on spider populations also have received scant attention, particularly in North America. Coyle (1981) studied the effects of clearcutting on the spider community of a southern Appalachian forest in North Carolina. The effects of silvicultural practices on European forest spiders have received more attention; studies include those by Huhta et al (1967, 1969) and Huhta (1971). As part of our investigations on natural enemies of the spruce budworm, Choristoneura fumiferana (Clem.), we studied the spider fauna of strip clem cur and dense (uncut) spruce-fir forests of northern Maine. in 1977 and 1978. Spruce bud worms are susceptible to ground-inhabiting predators when 1st and 2nd instars disperse (Mott 1963; Jennings et al 1983), and when large larvae and pupae drop from host-tree crowns to the forest floor (Morris and Mott 1963; Kelly and Regniere 1985). Our objectives were to: (1) determine the species of ground-inhabiting spiders in uncut residual strips, clearcut strips, and dense (uncut) spruce-fir stands, (2) determine seasonal activities of ground-inhabiting spiders as they relate to spruce budworm development, and (3) determine possible effects of strip clearcutting on species diversity and evenness of distribution of spiders. We also investigated effects of strip-clearcut age and litter depth on numbers and species of spiders. MATERIALS AND METHODS Study area. — We studied spiders in a dense spruce-fir forest infested with spruce budworm. Portions of the forest had been strip clearcut by mechanical harvesters; this created open strip areas with abundant shrubs and forbs, mainly Rubus spp. Strip clearcutting resulted in alternating clearcut and uncut residual strips (Fig. 1). Individual study sites were located from 48 to 61 km northwest of Millinocket, Piscataquis County, Maine, between Telos and Harrington Lakes (45° 45' to 46° 10' N, 68° 55' to 69° 20' W). Elevations ranged from about 360 to 425 m. The forest stands previously had been infested with spruce budworm for 4 to 5 years. The study area was sprayed with Sevin-4-oil® for spruce budworm suppression in 1976, but was not sprayed in 1977 or 1978. Budworm population estimates were 92.8 larvae-pupae/m2 of balsam-fir foliage in 1977, and 100.7 larvae-pupae/m2 of foliage in 1978. Five strip-clearcut stands and five nearby dense (uncut) stands were investigated in 1977; seven strip-clearcut stands and three dense stands were investigated in 1978. In 1978, three each of the strip-clearcut and dense stands were the same as those studied in 1977; four additional strip-clearcut stands were investigated to obtain information on possible effects of strip-clearcut age on spider populations. Strip widths ranged from 23.4 to 49.7 m for uncut residual JENNINGS ET AL.— SPIDERS AND SPRUCE-FIR FORESTS 57 Fig. 1. — Pitfall-trap layout in strip-clearcut (uncut residual and clearcut strips) and in nearby dense (uncut) spruce-fir stands, Piscataquis County, Maine. strips and from 19.1 to 29.7 m for clearcut strips investigated in 1977. Strip widths were not measured in 1978 but were comparable to those studied in 1977. Mean basal areas, tree heights, and stand ages in 1977 were: 33.1 m2/ha, 17.3 m, and 73.7 years for uncut residual strips; 41.5 m2/ha, 15.4 m, and 73.5 years for dense stands. Most study sites had a predominantly softwood component of balsam fir, Abies balsamea (L.) Mill.; red spruce, Picea rubens Sargent; white spruce, P. glauca (Moench) Voss; black spruce, P. mariana (Miller) B.S.P.; northern white cedar, Thuja occidentalis L.; and white pine, Pinus strobus L. Common hardwood species were paper birch, Betula papyrifera Marshall; yellow birch, B. alleghaniensis Britton; and red maple, Acer rubrum L. Species composition by percentage basal area indicated that dense stands had more spruce (80%) than fir (11%), whereas spruce (45%) and fir (43%) were about equally represented in the uncut residual strips studied in 1977. The additional uncut residual strips studied in 1978 had more spruce (61%) than fir (30%). Hardwood basal area percentages generally were < 5% for both dense stands and uncut residual strips. Understory vegetation differed markedly among the forest conditions. The open cleared strips had an abundance of flowering shrubs and forbs such as Kalmia angustifolia L., Prunus pensylvanica L., Vaccinium angustif olium Aiton, and Rubus spp. The uncut residual strips and dense stands, on the other hand, were characterized by few, widely spaced or clumped plants such as Maianthemum canadense Desfontaines, Oxalis montana Rafinesque-Schmaltz, and Cornus canadensis L. No quantitative plant data were taken; however, plants of 21 58 THE JOURNAL OF ARACHNOLOGY families, 43 genera, and at least 50 species were collected and identified (Jennings and Houseweart, unpublished data). Most plant species were collected in open areas of strip clearcuts; only five species were collected in uncut residual strips and 10 species in dense (uncut) stands. Pitfall traps. — Forty large-capacity pitfall traps (Houseweart et al. 1979) were used each year for collecting spiders. Although pitfall catches are biased toward active forms, pitfall trapping remains the best available means for sampling wandering spiders (Uetz 1975), and trap catches give a closer estimate of species diversity than quadrat sampling (Uetz and Unzicker 1976). Our large-capacity trap (1 liter) had a 30-cm2 apron with a 14.9-cm-diameter hole for funnel-bottle suspension. Cutler et al. (1975) showed that traps with aprons caught twice as many dionychous spiders compared to traps without aprons. We added ca. 300 ml of a 1:1 mixture of ethylene glycol and 70% ethanol to each trap bottle as a killing-preservative agent. Four traps were placed in each strip-clearcut area, one each in two cut strips and in two adjacent uncut residual strips (Fig. 1). Correspondingly, four traps were placed in each nearby dense (uncut) stand investigated. Trap spacings in strip clearcuts were duplicated in dense stands. Traps were installed on 26 May and their contents collected weekly thereafter for 10 weeks from 2 June to 4 August 1977. In 1978, traps were installed on 18 May and contents collected weekly thereafter for 1 1 weeks from 25 May to 3 August. For both study years, trapping periods corresponded with spruce budworm activity, i.e., larval feeding in May and June; pupation in late June; moth emergence, mating, and egg laying in mid July (Houseweart et al. 1982); and dispersal of 1st instars in late July (Jennings et al. 1983). Trap contents were sorted in the laboratory; all spiders were removed and stored in 2-dram neoprene-stoppered vials containing 70% ethanol. Species determinations were made chiefly after Kaston (1981). Other consulted sources included: Opell and Beatty (1976) for the hahniids; Leech (1972) for the amaurobiids; Chamberlin and Gertsch (1958) for the dictynids; Levi (1957) for species of Theridion ; Dondale and Redner (1982) for the clubionids; Dondale and Redner (1978) for the philodromids and thomisids; and Kaston (1973) for species of Metaphidippus. Numerous taxonomic papers were consulted for identification of the erigonids; most were identified by comparison with published species descriptions and with voucher specimens housed in the Canadian National Collection, Ottawa. A few adult erigonids could not be determined to species and were designated as sp. 1, sp. 2, etc. Because most species descriptions are based chiefly on the genitalia, only sexually mature spiders were identified to species. Juvenile and penultimate stages were identified to generic level; recently emerged spiderlings to family level. A few badly damaged specimens were undeterminable. Representative specimens of most collected spider species are deposited in the arachnid collections of the American Museum of Natural History, New York; the Canadian National Collections of Insects, Arachnids, and Nematodes, Ottawa; and, the U.S. National Museum of Natural History, Washington. Litter depth. — Because litter structure and depth significantly affect abundances of some forest floor spiders (Uetz 1979; Bultman et al. 1982), we measured litter depth (cm) near each pitfall trap ( n = 40). Measurements were summed and JENNINGS ET AL.— SPIDERS AND SPRUCE-FIR FORESTS 59 means calculated over all replications by forest stand condition (uncut residual strip, clearcut strip, dense stand). Data analysis. — Pitfall-catch data were subjected to Hartley’s Test for homogeneity of variance prior to statistical analyses. Natural log transformations, In (X + 1), were used to stabilize variances. Analysis of variance (ANOVA) and Duncan’s Multiple Range Test were used to evaluate differences in pitfall catches over all weeks among the three forest stand conditions (uncut residual strips, clearcut strips, and dense stands) at P < 0.05. Regression analyses were used to evaluate the effects of strip-clearcut age (1-6 yr) and litter depth (independent variables) on mean catches of both individuals and species (dependent variables) per trap per week, where R2 = coefficient of determination. Because our pitfall collections represented finite populations where all captured individuals were counted and identified (Pielou 1966; Poole 1974), we used Brillouin’s diversity formula to calculate species diversity. The formula as defined by Pielou (1975 p. 10) is: H = 1/N log Ni/UN.l where N is the number of individuals in the whole collection (i.e., for each forest condition) and N; is the number in the ith species for i = Brillouin’s formula has been used to compare pitfall-catch diversities of spiders (Doane and Dondale 1979), carabid beetles (Reeves et al. 1983), phalangids (Jennings et al. 1984), and ants (Jennings et al. 1986). A measure of evenness was determined by the formula J = H/Hmax where H is Brillouin’s diversity and Hmax is the maximum possible diversity. Two measures of similarity among forest conditions were made using coefficient of community (CC) and percent similarity (PS) (Pielou 1975), where (CC) measures similarity between species lists and (PS) measures similarity between species quantities. RESULTS Numbers of individuals and species. — Fully 11,107 spiders, representing 15 families, 76 genera, and at least 125 species were collected by pitfall traps in spruce-fir forests of northern Maine. Fifteen families, 62 genera, and at least 97 species were trapped in 1977; 15 families, 66 genera, and at least 105 species were trapped in 1978 (Table 1). Generic and species composition differed between years. Ten species were captured in 1977 but not in 1978; 14 species were trapped in 1978 but not in 1977. For both years of trapping, individuals of Lycosidae were numerically dominant, comprising 56.2 and 54.1% of the total specimens trapped in 1977 and 1978, respectively. The Erigonidae, Amaurobiidae, and Agelenidae were next in abundance; each of the remaining families accounted for less than 10% of the total spiders caught either year. Although the 40 pitfall traps were distributed unevenly among forest conditions, for both study years more spiders were captured in the clearcut strips than in either the uncut residual strips or in the dense spruce-fir stands. By far the majority of spiders caught in the open, cleared strips were species of Pardosa and undetermined lycosid spiderlings. Members of no other genus or family approached the abundance of wolf spiders in clearcut strips. For both study years, about equal numbers of spider species were collected in dense, uncut spruce-fir stands (Table 1). In 1978, more species of spiders were caught in clearcut strips than the other two forest conditions. 60 THE JOURNAL OF ARACHNOLOGY Table 1. — Species and numbers of spiders collected in pitfall traps, three forest conditions, Telos Lake area, Piscataquis County, Maine, 1977-1978 (C = dearcut strips; R = uncut residual strips; D = dense stands; N = number of pitfall traps) 1977 1978 Spider Species CRD Strips Strips Stands (#=10) (#=10) (#=20) Total % CRD Strips Strips Stands (#=14) (#=14) (#=12) Total % AGELENIDAE Agelenopsis utahana 1 3 4 8 0.20 22 23 11 56 0.79 Agelenopsis sp. 0 0 1 1 0.02 3 1 0 4 0.06 Cicurina brevis 7 9 17 33 0.83 10 26 24 60 0.85 Cicurina pallida 4 8 14 26 0.65 15 13 5 33 0.47 Cicurina placida 0 0 1 1 0.02 0 0 0 0 Cicurina sp. 4 2 4 10 0.25 10 3 1 14 0.20 Coras montanus 0 0 4 4 0.10 1 1 1 3 0.04 Coras sp. 0 1 1 2 0.05 0 0 0 0 Cryphoeca montana 1 12 19 32 0.80 6 85 51 142 2.00 Cryphoeca sp. 0 0 2 2 0.05 0 1 0 1 0.01 Wadotes calcaratus 5 43 105 153 3.81 33 121 102 256 3.61 Wadotes sp. 2 5 9 16 0.40 8 22 17 47 0.67 HAHNIIDAE Antistea brunnea 1 0 1 2 0.05 0 0 1 1 0.01 Hahnia cinerea 0 0 5 5 0.12 2 0 2 4 0.06 Hahnia sp. 0 0 1 1 0.02 0 0 0 0 Neoantistea agilis 0 0 0 0 3 0 1 4 0.06 Neoantistea magna 69 14 16 99 2.46 256 65 33 354 5.00 Neoantistea sp. 8 0 1 9 0.22 52 1 2 55 0.78 Undet. sp. 1 0 2 3 0.08 1 0 0 1 0.01 AMAUROBIIDAE Amaurobius borealis 8 42 134 184 4.58 33 42 18 93 1.32 Amaurobius sp. 2 0 2 4 0.10 0 0 0 0 Caliioplus euoplus 0 0 2 2 0.05 0 2 0 2 0.03 Callioplus tibialis 0 18 22 40 1.00 2 25 10 37 0.52 Caliioplus sp. i 0 0 1 0.02 0 2 1 3 0.04 Callobius bennetti 20 57 67 144 3.58 42 113 40 195 2.75 Cailobius sp. 12 37 36 85 2.11 17 43 25 85 1.20 Undet. sp. 3 4 5 12 0.30 7 4 2 13 0.19 DiCTYMIDAE Diclyna brevitarsus 1 0 0 1 0.02 0 0 0 0 Lathys pallida 0 0 2 2 0.05 0 0 1 1 0.01 Undet. sp. 0 1 2 3 0.08 0 0 1 1 0.01 THERIDIIBAE Robertus fuscus 0 2 2 4 0.10 1 0 2 3 0.04 Robertus riparius 17 2 11 30 0.75 14 6 5 25 0.36 Robertus sp. 2 0 2 4 0.10 2 0 3 5 0.07 Theonoe stridula 0 0 11 11 0.28 0 2 10 12 0.17 Theridion montanum 0 0 0 0 0 2 0 2 0.03 Theridion sexpunctatum 0 0 2 2 0.05 0 0 0 0 Theridion sp. 0 0 2 2 0.05 0 0 2 2 0.03 Undet. sp. 1 0 1 2 0.05 0 0 0 0 LINYPHIIDAE Agyneta olivacea 0 0 0 0 0 1 1 2 0.03 Aphiieta misera 1 0 0 1 0.02 1 0 0 1 0.01 Ralhyphantes paliidus 7 19 4 30 0.75 30 54 17 101 1.43 Bathyphantes sp. 1 0 1 2 0.05 0 0 0 0 Centromerus denticulatus 0 0 6 6 0.15 0 0 7 7 0.10 JENNINGS ET AL.— SPIDERS AND SPRUCE-FIR FORESTS Centromerus furcatus 4 11 16 31 0.78 4 8 7 19 61 0.27 Centromerus longibulbus 0 0 1 1 0.02 0 0 0 0 Centromerus persolutus 2 0 2 4 0.10 4 4 0 8 0.11 Lepthyphantes alpinus 0 16 19 35 0.88 3 37 50 90 1.27 Lepthyphantes complicatus 0 0 1 1 0.02 0 5 2 7 0.10 Lepthyphantes intricatus 2 13 3 18 0.45 6 14 4 24 0.34 Lepthyphantes sp. near arboreus 0 0 1 1 0.02 0 0 1 1 0.01 Lepthyphantes turbatrix 0 0 0 0 0 2 0 2 0.03 Lepthyphantes zebra 0 0 1 1 0.02 0 0 2 2 0.03 Lepthyphantes sp. 0 2 2 4 0.10 0 0 0 0 Meioneta simplex 0 0 0 0 4 0 0 4 0.06 Oreonetides flavescens 0 0 0 0 1 0 0 1 0.01 Oreonetides recurvatus I 0 0 1 0.02 2 1 0 3 0.04 Oreonetides rotundus 0 2 0 2 0.05 1 0 1 2 0.03 Oreonetides vaginatus 0 10 22 32 0.80 4 19 14 37 0.52 Oreonetides sp. 1 0 0 0 0 0 0 2 2 0.03 Oreonetides sp. 2 2 1 1 4 0.10 3 11 2 16 0.23 Porrhomma sp. 1 1 0 2 0.05 0 1 1 2 0.03 Wubana drassoides 0 2 1 3 0.08 0 2 0 2 0.03 Undet. sp. 2 1 1 4 0.10 10 18 24 52 0.74 ERIGONIDAE Baryphyma longitarsum 1 0 0 1 0.02 3 0 0 3 0.04 Carorita limnaeus 0 0 0 0 0 0 2 2 0.03 Ceraticelus atriceps 1 0 0 1 0.02 0 0 0 0 Ceraticelus fissiceps 0 0 0 0 1 0 0 1 0.01 Ceraticelus laetabilis 5 6 1 12 0.30 21 16 3 40 0.57 Ceraticelus minutus 2 0 0 2 0.05 5 0 0 5 0.07 Ceratinella brunnea 13 5 34 52 1.30 32 81 83 196 2.77 Ceratinella sp. 9 2 2 13 0.33 6 5 6 17 0.24 Ceratinopsis auriculata 0 0 0 0 0 1 0 1 0.01 Dicymbium elongatum 0 0 0 0 1 0 1 2 0.03 Diplocentria bidentata 8 29 15 52 1.30 26 42 28 96 1.36 Diplocentria rectangulata 0 0 2 2 0.05 0 0 1 1 0.01 Diplocephalus cuneatus 1 0 1 2 0.05 0 0 0 0 Eperigone entomologica 0 0 0 0 0 4 3 7 0.10 Eperigone maculata 9 1 0 10 0.25 10 5 3 18 0.25 Eperigone trilobata 67 3 0 70 1.74 83 5 0 88 1.24 Erigone atra 1 1 0 2 0.05 0 0 0 0 Erigone sp. 1 1 0 0 1 0.02 0 0 0 0 Erigone sp. 2 1 0 0 1 0.02 0 0 0 0 Floricomus plumalis 3 0 3 6 0.15 9 1 3 13 0.19 Gnathonaroides pedale 0 0 0 0 0 0 1 1 0.01 Gonatium crassipalpum 0 1 0 1 0.02 0 0 0 0 Grammonota angusta 0 2 2 4 0.10 0 6 7 13 0.19 Grammonota gigas 0 0 0 0 12 1 0 13 0.19 Grammonota sp. 0 0 1 1 0.02 4 2 3 9 0.13 Halorates sp. 0 1 4 5 0.12 0 0 2 2 0.03 Islandiana longisetosa 0 0 0 0 0 1 0 1 0.01 Oedothorax trilobatus 2 0 0 2 0.05 1 0 0 1 0.01 Pocadicnemis americana 13 11 76 100 2.49 87 27 100 214 3.01 Sciastes truncatus 1 2 5 8 0.20 0 0 8 8 0.11 Scironis tarsalis 4 0 0 4 0.10 3 0 0 3 0.04 Scotinotylus pallidus 0 0 0 0 0 1 0 1 0.01 Sisicottus montanus 2 11 21 34 0.85 7 48 12 67 0.95 Sisicus apertus 0 0 1 1 0.02 0 0 0 0 Sisicus penifusiferus 0 0 1 1 0.02 0 0 0 0 Tapinocyba bicarinata 0 3 3 6 0.15 1 0 6 7 0.10 62 Tapinocyba minuta 0 1 Tapinocyba simplex 0 3 Tunagyna debilis 10 4 Walckenaeria atrotibialis 0 0 Walckenaeria castanea 0 3 Walckenaeria communis 0 0 Walckenaeria directa 0 0 Walckenaeria exigua 14 7 Walckenaeria spiralis 1 0 Walckenaeria teres 0 1 Undet. sp. 8 5 ARANEIDAE Araneus nordmanni 0 0 Undet. sp. 1 0 MIMETIDAE Ero canionis 1 0 Ero sp. 1 0 LYCOSIDAE Alopecosa aculeata 6 0 Alopecosa sp. 0 0 Arctosa raptor 1 0 Hogna sp. 2 0 Pardosa fuscula 0 0 Pardosa hyperborea 3 1 Pardosa mackenziana 290 37 Pardosa maesta 145 1 Pardosa xerampelina 314 12 Pardosa sp. 93 3 Pirata cant r alii 2 0 Pirata insularis 6 0 Pirata minutus 9 0 Pirata montanus 0 0 Pirata piraticus 4 1 Pirata sp. 1 0 Trochosa terricola 6 2 Trochosa sp. 0 0 Undet. sp. 1133 113 GNAPHOSIDAE Gnaphosa parvula 2 0 Gnaphosa sp. 0 0 Haplodrassus signifier 1 0 Micaria pulicaria 0 2 Orodrassus canadensis 0 1 Zelotes fratris 9 0 Zelotes sp. 2 0 Undet. sp. 1 0 CLUBIONIDAE Agroeca ornata 0 1 Agroeca sp. 0 0 Clubiona bishopi 1 0 Clubiona canadensis 0 5 Clubiona sp. 3 0 PHILODROMIDAE Philodromus placidus 1 1 Philodromus sp. 0 0 Tibellus oblongus 0 0 THOMISIDAE Ozyptila distans 0 0 Xysticus canadensis 0 19 THE JOURNAL OF ARACHNOLOGY 2 0.05 7 6 1 14 0.20 16 0.40 6 14 15 35 0.49 17 0.42 18 20 11 49 0.69 2 0.05 7 0 0 7 0.10 8 0.20 0 1 0 1 0.01 0 1 0 1 2 0.03 1 0.02 2 0 6 8 0.11 37 0.92 8 7 11 26 0.37 1 0.02 4 0 0 4 0.06 1 0.02 0 0 0 0 41 1.02 23 12 12 47 0.67 0 1 0 0 1 0.01 3 0.08 3 2 1 6 0.06 2 0.05 0 0 1 1 0.01 1 0.02 0 0 0 0 6 0.15 11 1 1 13 0.19 4 0.10 0 0 0 0 1 0.02 0 0 0 0 2 0.05 0 0 0 0 0 1 0 0 1 0.01 9 0.22 22 0 3 25 0.36 337 8.38 504 91 24 619 8.74 147 3.66 332 2 0 334 4.71 326 8.11 770 75 7 852 12.03 101 2.51 166 8 3 178 2.51 2 0.05 0 0 0 0 15 0.37 7 0 4 11 0.16 10 0.25 25 0 0 25 0.36 0 2 0 1 3 0.04 5 0.12 1 0 0 1 0.01 3 0.08 3 0 3 6 0.09 42 1.05 37 21 53 111 1.57 0 5 3 3 11 0.16 1248 31.04 1368 222 56 1646 23.23 2 0.05 7 0 0 7 0.10 0 1 0 0 1 0.01 1 0.02 2 0 1 3 0.04 2 0.05 1 0 0 1 0.01 1 0.02 0 0 0 0 11 0.28 27 0 2 29 0.41 2 0.05 18 1 0 19 0.27 1 0.02 1 0 0 1 0.01 9 0.22 1 6 11 18 0.25 2 0.05 0 0 1 1 0.01 1 0.02 0 0 0 0 13 0.33 0 8 2 10 0.14 4 0.10 0 7 2 9 0.13 3 0.08 1 2 1 4 0.06 0 0 1 1 2 0.03 0 1 0 0 1 0.01 0 0 0 1 1 0.01 54 1.34 2 88 95 185 2.61 1 13 3 2 5 0 1 16 0 0 28 0 2 1 0 0 4 0 0 0 5 10 1 0 5 0 9 1 0 0 2 34 0 2 0 0 0 0 0 2 0 0 8 2 0 8 1 1 0 0 0 35 JENNINGS ET AL.— SPIDERS AND SPRUCE-FIR FORESTS 63 Xysticus discursans 0 0 0 0 1 0 0 1 0.01 Xysticus emertoni 0 0 0 0 7 0 0 7 0.10 Xysticus sp. 0 0 2 2 0.05 0 2 2 4 0.06 SALTICIDAE Euophrys cruciatus 4 0 1 5 0.12 0 0 0 0 Evarcha hoyi 0 0 0 0 2 0 0 2 0.03 Metaphidippus flavipedes 0 0 1 1 0.02 0 1 2 3 0.04 Pellenes montanus 0 0 0 0 4 0 0 4 0.06 Phidippus borealis 0 0 0 0 3 0 0 3 0.04 Phidippus whitmanii 0 0 0 0 1 0 0 1 0.01 Sitticus finschii 0 0 1 1 0.02 0 0 0 0 Undet. sp. 0 0 0 0 0 0 1 1 0.01 UNIDENTIFIABLE 0 0 0 0 1 0 0 1 0.01 Subtotals: Species 59 51 68 76 56 69 Individuals 2412 639 971 4022 4340 1627 1118 7085 Species of spiders that showed habitat affinities both study years were: Neoantistea magna (Keyserling), Eperigone trilobata (Emerton), Pardosa macken - ziana (Keyserling), P. moesta Banks, P. xerampelina (Keyserling), Pirata minutus Emerton, and Zelotes fratris Chamberlin for clearcut strips; Bathyphantes pallidus (Banks) and Diplocentria bidentata (Emerton) for uncut residual strips; Wadotes calcaratus (Keyserling), Callobius bennetti (Blackwall), Lepthyphantes alpinus (Emerton), Oreonetides vaginatus (Thorell), Ceratinella brunnea Emerton, Sisicottus montanus (Emerton), and Xysticus canadensis Gertsch for uncut residual strips, dense stands, or both uncut forest conditions. Habitat preferences of Pocadicnemis americana Millidge and Trochosa terricola Thorell were undeterminable from our data. In 1977 and 1978, both species were most abundant in dense stands; however, in 1978, both species were also prevalent in clearcut strips. Pitfall catches for the remaining species showed no clear habitat preference or affinity. Mean numbers of individuals and species. — For both study years, significantly more individuals (ANOVA, F = 84.7, P < 0.0001, 1977; F = 41.7, P < 0.0001, 1978) and more species (ANOVA, F = 32.3, P < 0.0001, 1977; F = 13.4, P < 0.0001, 1978) of spiders were captured in clearcut strips than in either uncut residual strips or dense stands (Table 2). And, for both study years, the most similar habitats (i.e., uncut residual strips and dense stands) did not differ significantly for mean numbers of individuals per trap per week. Likewise, mean numbers of species did not differ significantly either study year. Table 2. — Mean numbers of individuals and spider species per trap per week by forest condition. Column means followed by the same letter are not significantly different at the P < 0.05 level, Duncan’s multiple range test. Natural log transformations, In (x + i), were used to stabilize variances of mean numbers of individuals. X (± SE) individuals X (+ SE) species Forest condition 1977 1978 1977 1978 Clearcut strips Residual strips Dense stands 24.12a (2.21) 6.39b (0.70) 4.85b (0.33) 28.18a (2.04) 10.56b (0.80) 8.48b (0.64) 5.98a (0.27) 3.94b (0.29) 3.33b (0.19) 7.40a (0.35) 5.77b (0.31) 5.06b (0.32) 64 THE JOURNAL OF ARACHNOLOGY Fig. 2. — Mean catches of spiders per trap per week; week 1 = 2 June 1977. Open triangles = dense spruce-fir stands; open circles = uncut residual strips; open squares = clearcut strips. Seasonal activity. — Spiders were active during most of the spruce budworm’s developmental stages; however, seasonal activities varied between study years and among forest conditions (Figs. 2 and 3). For both study years, mean catch densities per week were greater in clearcut strips; whereas, mean catches were about equal for uncut residual strips and dense stands. In both 1977 and 1978, peak catches were observed the 3rd and 6th weeks for clearcut strips; the latter peaks generally corresponded with emergence of young spiderlings. Mean catches of individuals per week generally declined about the 3rd (1977) and 2nd (1978) weeks of trapping for both dense stands and uncut residual strips. Mean species per trap also fluctuated between years and among forest conditions (Figs. 4 and 5). Strip clearcuts generally had more species per trap per week than the other two forest conditions. For both study years, mean species catch rate generally declined after the 5th and 6th weeks of trapping; however, in 1978 catch rates increased during the 10th and 11th weeks. Species diversity and evenness. — Although more individuals and species of spiders were captured in clearcut strips, species diversity and evenness of pitfall- trap catches were generally greater in dense stands and uncut residual strips (Table 3). The most similar habitats (i.e., uncut residual strips and dense stands) had comparable diversity and evenness values both years. No doubt the uneven distribution of individuals in clearcut strips contributed to lower species diversity for these habitats. Species diversity increases as individuals become more evenly distributed (Price 1975). The higher observed variances in mean individuals per trap per week (Table 2) also indicate unevenly distributed individuals for clearcut- strip habitats. Coefficient of community and percentage similarity. — Strip-clearcut areas (uncut residual and clearcut strips) and dense stands shared about the same number of species of spiders either study year; CC = 68.1 in 1977; CC = 66.2 in 1978. Surprisingly, the most dissimilar neighboring habitats (i.e., uncut residual and clearcut strips) had only slightly fewer species in common; CC = 58.7 in 1977; CC = 62.1 in 1978. JENNINGS ET AL.— SPIDERS AND SPRUCE-FIR FORESTS 65 Fig. 3. — Mean catches of spiders per trap per week; week 1 = 25 May 1978. Open triangles = dense spruce-fir stands; open circles = uncut residual strips; open squares = clearcut strips. Among strip-clearcut areas and dense stands, percentage similarity in numbers of individuals of each spider species was about the same both study years, i.e., PS = 39.6 in 1977; PS = 39.9 in 1978. However, comparing catches within strip clearcuts (uncut residual vs. clearcut strips) showed that fewer individuals shared these habitats in common in 1977 (PS = 22.8) than in 1978 (PS == 32.8). The relatively low PS values support our hypothesis that the unevenly distributed individuals in clearcut strips contributed greatly to lower species diversity for these habitats. Age of strip clearcut. — Regression analyses indicated that age of strip clearcut (1=6 years) had little influence on mean catches of individuals (R2 = 0.08, P > 0.37) and mean numbers of species (R2 = 0.41, P > 0.02) of spiders/ trap/ week. Litter depth. — Mean litter depth was significantly greater (P < 0.01) in dense stands (X = 11.7 cm) than in either uncut residual (X = 8.5 cm) or in clearcut strips (X — 8.0 cm). Uncut residual and clearcut-strip means did not differ significantly. Regression analyses indicated that litter depth had little influence on either mean catches (R2 = 0.04, P > 0.08) or mean species (R2 = 0.05, P > 0.05) of spiders/ trap/ week. DISCUSSION Few previous studies have dealt with Maine spiders. The spider fauna of Mount Desert Island, Hancock County, has been studied most extensively; Procter (1946) listed 15 families, 94 genera, and 179 species from various habitats on the island. Earlier, Bishop and Clarke (1923) reported spiders of 10 families, 23 genera, and 29 species from Isle au Haut, Knox County, Maine. Blake (1926) studied the biota of Mount Katahdin, Piscataquis County, and reported the collection of 20 species, which represent 14 genera and 11 families of spiders. Mount Katahdin is about 25 km southeast of our spruce-fir study area. Our collections of spiders from spruce-fir forests in Piscataquis County substantially adds to the species recorded from Maine. Our results on mean catches of individuals and species per trap per week generally indicate that the ground-inhabiting spiders preferred the more open, 66 THE JOURNAL OF ARACHNOLOGY Fig. 4. — Mean species of spiders per trap per week; week 1 = 2 June 1977. Open triangles = dense spruce-fir stands; open circles = uncut residual strips; open squares = clearcut strips. cleared habitats of clearcut strips to that of closed, shaded habitats of uncut residual strips and dense stands. Significantly more individuals and species were caught in the clearcut strips both study years (Table 2). However, most of this unequal distribution can be attributed to the greater abundance of lycosid spiders, and particularly Pardosa species, in the clearcut strips. For both study years, significantly more (P < 0.05) individuals of Pardosa species were trapped in the clearcut strips than in the other forest conditions, X = 8.4/ trap/ week in 1977 and X= 1 1.7/ trap/ week in 1978. Measures of similarity among forest conditions indicated that more species (CC values) than individuals (PS values) shared forest conditions in common. No doubt heterogenity of habitats and individual species requirements influenced the species composition and spatial distributions of spiders among the forest conditions studied. Species such as Neoantistea magna , Pardosa mackenziana , and P. xerampelina showed definite habitat affinities for clearcut strips; other species, such as Wadotes calcaratus , Callobius bennetti , Lepthyphantes alpinus , and Xysticus canadensis were intermediate in habitat association (i.e., two forest conditions, both similar); whereas, Pocadicnemis americana and Trochosa terricola were indeterminate regarding habitat preference. Because species diversity and evenness of pitfall trap-catches were generally lower in clearcut strips both study years, we conclude that strip clearcuttieg may alter species richness and abundance of ground-dwelling spiders in northeastern spruce-fir forests. The open, cleared strips with abundant shrubs and forbs provide new and different habitats where hunting spiders (e.g., Lycosidae) abound. Coyle (1981) also found an abundance of hunting spiders in clearcuts of a southern Appalachian forest. The uncut residual stands of northern Maine, conversely, provide islands of “refugia” where spider populations are diverse. Thus, the overall effect of strip clearcutting (clearcut and uncut residual strips) is a reduction in species diversity and evenness of spiders (Table 3), but an increase in spider abundance (Table 2). With time and plant succession, the strip -clearcut areas should provide macrohabitats and microhabitats similar to dense stands. JENNINGS ET AL.— SPIDERS AND SPRUCE-FIR FORESTS 67 Fig. 5. — Mean species of spiders per trap per week; week 1 = 25 May 1978. Open triangles = dense spruce-fir stands; open circles = uncut residual strips; open squares = clearcut strips. 25 I 8 15 22 29 6 13 20 27 3 May I June I July I Aug. WEEK We detected little influence of either strip-clearcut age or litter depth on mean overall catches of individuals and species of spiders. However, strip-clearcut age may become increasingly important during later successional stages when forest regeneration causes canopy closure. Canopy closure, or absence thereof, apparently had a dramatic effect on the abundance and distribution of certain Lycosidae (e.g., species of Pardosa ) in the northeastern spruce-fir forest. Most of the wolf spiders were found in open, sunny areas of clearcut strips, with lesser numbers in closed, shaded areas of dense stands and uncut residual strips. Bultman et al. (1982) also found a scarcity of wolf spiders in a beech-maple climax forest of western Michigan. Several investigators (Hagstrum 1970; Uetz 1979; Bultman et al. 1982) have concluded that litter depth is an important factor that influences spider abundance, particularly in deciduous forests. We suspect that litter depth may have lesser influence on spider abundance in coniferous forests where litter structure tends to be fairly uniform. Absence of canopy closure probably has a more significant effect on overall spider abundance in spruce-fir forests, and particularly on the abundance of Pardosa species. Our results on species diversity (Table 3) and litter depth generally agree with Uetz (1979); i.e., species diversity increased with increased litter depth of uncut dense stands. Based on pitfall-catch densities, spiders were abundant and active during the spruce budworm’s early and late larval stages, and during budworm oviposition and dispersal of first instars (weeks 8-10). Spruce budworm oviposition generally spans 27 days in late June and July (Houseweart et al. 1982), and budworm eggs are susceptible to predation by arboreal spiders (Jennings and Houseweart 1978). After budworm eggs hatch (ca. 10 days), the young larvae disperse and seek overwintering sites. However, during dispersal many budworms are lost because the larvae land on nonhost vegetation (Morris and Mott 1963); they also are susceptible to predation by spiders (Mott 1963). Strip-clearcutting contributed to dispersal losses of early-instar budworms (Jennings et al. 1983). Results of the current study indicate that budworm larve landing in clearcut strips would be especially susceptible to attack by lycosid spiders because of the latters’ great abundance. Predation on dropping late-instar budworms may also be significant 68 THE JOURNAL OF ARACHNOLOGY Table 3. — Spider species diversity and evenness of pitfall-trap catches by forest condition, Brillouin’s formula, 1977 and 1978. 1977, 5 replicates each forest condition; 4 traps/ replicate. 1978, 7 replicates (residual vs. clearcut strips); 3 replicates (residual + clearcut strips vs. dense stands); 4 traps/ replicate. Forest condition Diversity Evenness 1977 1978 1977 1978 Residual strips 1.31 1.37 0.81 0.81 Clearcut strips 0.98 1.08 0.57 0.58 Residual + clearcut strips 1.23 1.14 0.67 0.63 Dense stands 1.33 1.36 0.76 0.77 because mortality during the late larval stage influences generation survival of the spruce budworm (Morris 1963). Additional studies are needed to determine the predatory roles of spiders in northeastern spruce-fir forests and their impacts on spruce budworm populations. Such studies will require assessment of both predator and prey densities, and specialized techniques, such as the ELISA assay (Fichter and Stephen 1979; Ragsdale et al. 1981), to determine numbers of spruce budworms eaten by spiders. Enhancement of predator populations through forest management procedures is receiving increased attention (Crawford and Titterington 1979; Jennings and Crawford 1985). The current study indicates that populations of ground-dwelling spiders can be greatly increased by strip clearcutting; however, the long-term effects of strip clearcutting on changes in spider species diversity are unknown. Much more information is needed to develop sound, realistic pest management systems that promote and utilize natural agents of mortality and rely less on chemical insecticides. ACKNOWLEDGMENTS We thank our former technicians for field and laboratory assistance: Deirdre M. Haneman, Janice E. Littlefield, Susan L. May, Susan M. Sheffer, Denise E. Stephens, Michael Wissenbach, and Wesley A. Wright. Special thanks are due David M. Kendall for preparation of data summaries, Richard A. Hosmer for computer programming assistance, and Janet J. Melvin for word processing. Dr. Norman I. Platnick, American Museum of Natural History, New York, confirmed our determinations of Gnaphosidae. Drs. John Barron, Jonathan Coddington, Matthew H. Greenstone, and Gail E. Stratton provided constructive comments on an earlier draft of this paper. We are grateful to the Great Northern Paper Company, Woodlands Department, Millinocket, Maine, for permission to conduct these studies on their lands. This research paper is a contribution to the Canada/ United States (CANUSA) Spruce Budworms Program. LITERATURE CITED Bishop, S. C. and N. T. Clarke. 1923. A scientific survey of Turner’s Lake Isle au Haul, Maine. New York State Mus., Albany, New York, Aug. 1923 (published privately):21-27. JENNINGS ET AL.— SPIDERS AND SPRUCE-FIR FORESTS 69 Blake, I. H. 1926. A comparison of the animal communities of coniferous and deciduous forests. Illinois Biol. Monogr., 10(4): 1-149. Bultman, T. L., G. W. Uetz and A. R. Brady. 1982. A comparison of cursorial spider communities along a successional gradient. J. Arachnol., 10:23-33. Chamberlin, R. V. and W. J. Gertsch. 1958. The spider family Dictynidae in America north of Mexico. Amer. Mus. Nat. Hist. Bull., 116. 152 pp. Coyle, F. A. 1981. Effects of clearcutting on the spider community of a southern Appalachian forest. J. Arachnol., 9:285-298. Crawford, H. S. and R. W. Titterington. 1979. Effects of silvicultural practices on bird communities in upland spruce-fir stands. Pp. 110-119, In Management of North Central and Northeastern Forests for Nongame Birds. (R. M. DeGraaf and K. E. Evans, compilers). USDA For. Serv. Gen. Tech. Rep., NC-51. 268 pp. Cutler, B., L. H. Grim and H. M. Kulman. 1975. A study in the summer phenology of dionychious spiders from northern Minnesota forests. Great Lakes Entomol., 8:99-104. Doane, J. F. and C. D. Dondale. 1979. Seasonal captures of spiders (Araneae) in a wheat field and its grassy borders in central Saskatchewan. Canadian Entomol., 111:439-445. Dondale, C. D. and J. H. Redner. 1978. The crab spiders of Canada and Alaska (Araneae: Philodromidae and Thomisidae). Canadian Dep. Agric. Publ., 1663. 255 pp. Dondale, C. D. and J. H. Redner. 1982. The sac spiders of Canada and Alaska (Araneae: Clubionidae and Anyphaenidae). Canadian Dep. Agric. Publ., 1724. 194 pp. Dowdy, W. W. 1950. A community study of an oak-hickory association with special reference to invertebrates. Amer. Midland Nat., 43:667-695. Elliott, F. R. 1930. An ecological study of the spiders of the beech-maple forest. Ohio J. Sci., 30:1-22. Fichter, B. L. and W. P. Stephen. 1979. Selection and use of host-specific antigens. Pp. 25-33, In Serology in Insect Predator-prey Studies. (M. C. Miller, ed.). Misc. Pubis. Entomol. Soc. Amer. 11.84 pp. Gertsch, W. J. 1979. American Spiders. 2nd ed. Van Nostrand Reinhold Co., New York. 272 pp. Gibson, W. W. 1947. An ecological study of the spiders of a river terrace forest in western Tennessee. Ohio J. Sci., 47:38-44. Hagstrum, D. W. 1970. Ecological energetics of the spider Tarentula kochi. Ann. Entomol. Soc. Amer., 63:1297-1304. Haynes, D. L. and P. Sisojevic. 1966. Predatory behavior of Philodromus rufus Walckenaer (Araneae: Thomisidae). Canadian Entomol., 98:113-133. Houseweart, M. W., D. T. Jennings and J. C. Rea. 1979. Large capacity pitfall trap. Entomol. News, 90:51-54. Houseweart, M. W., S. G. Southard and D. T. Jennings. 1982. Availability and acceptability of spruce bud worm eggs to parasitism by the egg parasitoid, Trichogramma minutum (Hymenoptera: Trichogrammatidae). Canadian Entomol., 114:657-666. Huhta, V. 1971. Succession in the spider communities of the forest floor after clear-cutting and prescribed burning. Ann. Zool. Fennici, 8:483-542. Huhta, V., E. Karppinen, M. Nurminen and A. Valpas. 1967. Effect of silvicultural practices upon arthropod, annelid and nematode populations in coniferous forest soil. Ann. Zool. Fennici, 4:87- 143. Huhta, V., M. Nurminen and A. Valpas. 1969. Further notes on the effect of silvicultural practices upon the fauna of coniferous forest soil. Ann. Zool. Fennici, 6:327-334. Jennings, D. T. and M. W. Houseweart. 1978. Spider preys on spruce budworm egg mass. Entomol. News, 89:183-186. Jennings, D. T., M. W. Houseweart and J. B. Dimond. 1983. Dispersal losses of early-instar spruce budworm (Lepidoptera: Tortricidae) larvae in strip clearcut and dense spruce-fir forests of Maine. Environ. Entomol., 12:1787-1792. Jennings, D. T., M. W. Houseweart and J. C. Cokendolpher. 1984. Phalangids (Arachnida: Opiliones) associated with strip clearcut and dense spruce-fir forests of Maine. Environ. Entomol., 13:1306- 1311. Jennings, D. T., M. W. Houseweart and A. Francoeur. 1986. Ants (Hymenoptea: Formicidae) associated with strip-clearcut and dense spruce-fir forests of Maine. Canadian Entomol., 118:43-50. Jennings, D. T. and H. S. Crawford, Jr. 1985. Predators of the spruce budworm. U.S. Dep. Agric., Agric. Handb. 644. 77 pp. 70 THE JOURNAL OF ARACHNOLQGY Kaston, B. J. 1973. Four new species of Metaphidippus , with notes on related jumping spiders from the eastern and central United States. Amer. Microsc. Soc. Trans., 92:106-122. Kaston, B. J. 1981. Spiders of Connecticut. Bull. Connecticut State Geol. Nat. Hist. Surv., 70. 1020 pp. Kelly, B. and J. Regeiere. 1985. Predation on pupae of the spruce budworm (Lepidoptera: Tortricidae) on the forest floor. Canadian Entdmol, 117:33-38,. Leech, R. 1972. A revision of the nearctic Amaurobiidae (Arachnida: Araneida). Mem. Entomol. Soc. Canada, 84. 182 pp: Levi, H. W. 1957. The spider genera Enoplognatha , Theridion , and Paidisca in America north of Mexico. Amer. Mus. Nat, Hist. Bull., 112:5-123. Morris, R. F,, ed. 1963. The dynamics of epidemic spruce budworm populations. Mem. Entomol. Soc. Canada, 31. 332 pp. Morris, R. F. and D. G. Mott. 1963. Dispersal and the spruce budworm. Pp. 180-189, In The Dynamics of Epidemic Spruce Budworm Populations. (R. F. Morris, ed.). Mem. Entomol. Soc, Canada, 31. 332 pp. Mott, D. G. 1963. The analysis of the survival of small larvae in the unsprayed area. Pp. 42-52, In The Dynamics of Epidemic Spruce Budworm Populations. (R. F. Morris, ed.). Mem. Entomol. Soc. Canada, 31. 332 pp. Opell, B. D. and J. A. Beatty. 1976. The nearctic Hahniidae (Arachnida: Araneae). Bull. Mus. Comp. ZooL, !47(9):393-433. Pidou, E. C. 1966. The measurement of diversity in different types of biological collections. J. Them: . Biol, 13:131-144. Pielou, E. C. 1975. Ecological Diversity. John Wiley & Sons, New York. 165 pp. Poole, R. W. 1974. An Introduction to Quantitative Ecology. McGraw Hill, New York. 532 pp. Price, P. W. 1975. Insect Ecology. John Wiley & Sons, New York. 514 pp. Procter, W. 1946. Biological Survey of the Mount Desert Region Incorporated. Part VII. The Insect Fauna. The Wistar Institute of Anatomy and Biology, Philadelphia, PA. 566 pp. Ragsdale, D. W., A. D. Larson and L. D. Newsom. 1981. Quantitative assessment of the predators of Nezara viridula eggs and nymphs within a soybean agroecosystem using an ELISA. Environ. Entomol, 10:402-405. Reeves, R. M., G. A. Dunn and D. T. Jennings. 1983. Carabid beetles (Coleoptera: Carabidae) associated with the spruce budworm, Choristoneura fumiferana (Lepidoptera: Tortricidae). Canadian Entomol, 115:453-472. Riechert, 8. E. 1974. Thoughts on the ecological significance of spiders. BioScience, 24(6):352-356. Stratton, G. E., G. W. Uetz and D. G. Dillery. 1979. A comparison of the spiders of three coniferous tree species. J. Arachnol, 6:219-226. Uetz, G. W. 1975. Temporal and spatial variation in species diversity of wandering spiders (Araneae) in deciduous forest litter. Environ. Entomol., 4:719-724. Uetz, G. W. 1979. Influence of variation of litter habitats on spider communities. Oecologia, 40:29-42. Uetz, G. W. and J. D. Unzicker. 1976. Pitfall trapping in ecological studies of wandering spiders. J. Arachnol., 3:101-1 11. Manuscript received March 1987 , revised June 1987. Maury, E. A. 1988. Triaenonychidae Sudamericanos. III. Description de los nuevos generos Nahuelonyx y Valdivionyx (Opiliones, Laniatores). J. Arachnol., 16:71-83. TRIAENONYCHIDAE SUDAMERICANOS. III. DESCRIPCION DE LOS NUEVOS GENEROS NAHUELONYX Y VALDIVIONYX (OPILIONES, LANIATORES) Emilio A. Maury Museo Argentine de Ciencias Naturaies Angel Gallardo 470 (1405) Buenos Aires, Argentina ABSTRACT Two new genera and a new species of Triaenonychidae are described from the valdivian wet forest of southern Argentina and Chile: Nahuelonyx, new genus, for N. nasutus (Ringuelet 1959), new combination and Valdivionyx , new genus, for V crassipes , new species. The remarkable similarity in coloration of both species is mentioned. RESUMEN Dos nuevos generos y una nueva especie de Triaenonychidae se describen del bosque humedo valdiviano del sur de la Argentina y Chile: Nahuelonyx, genero nuevo, para N. nasutus (Ringuelet 1959), combination nueva y Valdivionyx, genero nuevo, para V crassipes, especie nueva. Se menciona la llamativa similaridad de coloration en ambas especies. INTRODUCCION En un reciente trabajo sobre el genero Diasia Sorensen 1902 (Maury, en prensa, a) mencione que la especie que Ringuelet (1959) describio como Diasia nasuta deberia ser ubicada en otro genero. Un examen mas minucioso de la serie tipica de dicha especie (cinco ejemplares) me revelo la existencia de dos entidades distintas, muy differentes a Diasia. Ambas pertenecen a la tribu Triaenonychini Sorensen 1902 y a mi juicio y de acuerdo a la literatura consultada, constituyen dos generos nuevos para la ciencia, que he denominado Nahuelonyx y Valdivionyx. Abundante material de ambas formas proveniente del sur de la Argentina y Chile me ha permitido efectuar adecuadas diagnosis, en donde se consignan los caracteres que a mi parecer son de importancia sistematica; las variaciones intraespecificas y esbozar un mapa de distribution geografica. Ambos generos son por el momento monotipicos y muy bien diferenciados por varios caracteres morfologicos y de genitalia, como puede verse en la clave adjunta. Lo que resulta sorprendente en dos generos tan bien caracterizados es la singular semejanza en la coloration. En los opiliones Triaenonychidae no se le ha dado mucha importancia a este caracter, pero en este caso merece un somero comentario, por las implicaciones de orden biologico que puedan tener. Salvo 72 THE JOURNAL OF ARACHNOLOGY minimos detalles diferenciales, Nahuelonyx y Valdivionyx presentan un color castano muy oscuro con veteado amarillento muy esfumado en cuerpo y apendices; pero como caracter distintivo, muestran dos resaltantes manchas ocelares amarillo ocre en el prosoma, a los costados del tuberculo ocular. Por observaciones personales realizadas en el bosque hurnedo valdiviano he comprobado que a diferencia de otros triaenoniquidos de la zona, de coloracion criptica, Nahuelonyx y Valdivionyx (a veces se los encuentra en simpatria) son facilmente detectables por estas manchas en los lugares donde habitan: reverso de troncos caidos o de cortezas y mantillo vegetal. Es de sospechar que dichas manchas cumplen alguna funcion. Pero como nada se conoce de la biologia de estos opiliones, solo se pueden hacer conjeturas sobre si la coloracion similar y tan llamativa de estos dos generos se debe a un parecido fortuito, a una evolution paralela o a algun caso de mimetismo. La nomenclatura utilizada en la genitalia masculina es fundamentalmente la de un trabajo anterior (Maury y Roig Alsina 1985), pero a partir del presente articulo se emplea la traduction espanola de algunos terminos: estilo (por stylus), laminilla (por lamella) y sensilo (por sensilia). CLAVE PARA DIFERENCIAR DIAS I A, NAHUELONYX Y VALDIVIONYX 1. Coloracion general castano amarillenta con manchas castano oscuro; en el prosoma no hay manchas que se destaquen. Prosoma mas largo que el escudo tergal. Borde anterior del prosoma sin hilera de tuberculos. Tuberculo ocular levemente conico o romo, sin apofisis terminal. Metatarsos de las patas I a IV sin separacion astragalo/calcaneo Diasia Coloracion general castano muy oscura con jaspeado amarillento; dos manchas ocelares amarillo ocre en el prosoma. Prosoma igual o mas corto que el escudo tergal. Borde anterior del prosoma con una hilera de tuberculos. Tuberculo ocular marcadamente conico, con una apofisis terminal. Metatarsos de las patas I a IV con separacion astragalo/calcaneo 2 2. Metatarsos I a IV con el calcaneo mayor que el astragalo. Tarsos III y IV similares en los dos sexos. Operculo genital grande, de forma semicircular. Estigmas respiratorios parcialmente ocluidos. Tuberculo ocular con una larga apofisis terminal. Areas del escudo tergal con gruesas granulaciones. Pene indicado en las Figs. 11 a 13. Ovipositor con dos apofisis curvas laterales Nahuelonyx Metatarso I con el astragalo igual que el calcaneo; metatarso II con el astragalo menor que el calcaneo; metatarsos III y IV con el astragalo mayor que el calcaneo. Tarsos III y IV del macho mas engrosados que en la hembra. Operculo genital pequeno, de forma ligeramente triangular. Estigmas respiratorios libres. Tuberculo ocular con corta apofisis terminal. Areas del escudo tergal con pequenas granulaciones. Pene indicado en las Figs. 16 a 18. Ovipositor sin apofisis laterales. Valdivionyx MAURY-DESCRIPCION de nahuelonyx and valdivionyx 73 Nahuelonyx , genero nuevo Diasia: Ringuelet 1959:256 (en parte, no Diasia Sorensen 1902). Especie tipo. — Nahuelonyx nasutus (Ringuelet 1959), por monotipia. Etimologia. — El nombre generico Nahuelonyx proviene de las palabras Nahuel , que en lengua indigena mapuche significa tigre y del griego onyx: una. Distribution. — Argentina: provincias de Neuquen y Rio Negro; Chile: provincias de Cautin, Valdivia, Osorno, Llanquihue, Chiloe y Palena (Figs. 32- 33). Diagnosis y description. — Triaenonychinae. Triaenonychini. Prosoma mas corto que el escudo tergal. Tuberculo ocular elevado, conico y con una aguda apofisis terminal. Prosoma con una hilera de tuberculos en todo el perimetro. Areas del escudo tergal poco definidas, inermes, con series transversales de gruesas granulaciones. Estigmas respiratorios parcialmente ocluidos por los tuberculos digitiformes del borde posterior de la coxa IV. Femures de las patas I a IV granulosos pero sin tuberculos que se destaquen. Operculo genital de borde libre semicircular, ligeramente mas ancho que largo en ambos sexos. Coxas I a IV con algunos pequenos tuberculos ventrales. Metatarsos I a IV con la separation astragalo / calcaneo bien marcada por una zona anular deprimida e incolora; en todos los metatarsos, aunque en diferente proportion, el calcaneo es mayor que el astragalo. En ambos sexos los tarsitos de la pata I ligeramente engrosados; en las otras patas normales. Distitarso de la pata I con dos segmentos; de la pata II con tres segmentos. Formula tarsal similar en los dos sexos: 3-7/11-4-4. Segmento basal de los queliceros con una pequena apofisis terminal dorsomedial. Dimorfismo sexual poco marcado: el operculo genital es proporcionalmente mas ancho que largo en la hembra que en el macho; en este ultimo sexo hay dos areas paramedianas postoperculares lisas y amarillentas, que no se ven en la hembra. Ovipositor bilobulado, con dos fuertes apofisis curvas laterales; hay cinco pares de sensilos ventrales y tres pares dorsoapicales. Pene con el estilo curvado hacia ventral; parte dorsolateral fuertemente bifurcada en su extreme distal, con un par de sensilos ventrales y una apofisis en la cara externa; la parte ventral lleva una laminilla hendida longitudinalmente, cuatro pares de sensilos y una apofisis lateral. Coloration general castano muy oscuro con jaspeado amarillento; una gran mancha ocelar amarilla a cada lado del prosoma. Nahuelonyx es, segun la literatura consultada, el unico triaenoniquido que posee en todos los metatarsos el astragalo menor que el calcaneo. Este caracter, sumado a la coloration distintiva (solo comparable a Valdivionyx) y a la peculiar morfologia de la genitalia, permitiran individualizar a este nuevo genero. Nahuelonyx nasutus (Ringuelet 1959), combination nueva Figs. 1-15, 33 Diasia nasuta Ringuelet 1959:259-263 (en parte); figs. 32 a-b, 33 a-b. Material tipico — Holotipo macho (MLP 24202) y paratipo hembra (MLP 24195): Lago Frias, Provincia de Rio Negro, Argentina. El alotipo hembra y dos paratipos macho y hembra de la serie tipica de Diasia nasuta pertenecen en realidad a Valdivionyx crassipes , genero y especie nuevos. 74 THE JOURNAL OF ARACHNOLOGY Figs. 1 = 10. — Nahuelonyx nasutus (Ringuelet): 1-8, macho de Pirehueico; 1, cuerpo, vista lateral; 2, region coxoesternal (detalle); 3, quelicero derecho, vista lateral; 4, pedipalpo derecho, vista lateral; 5, metatarso y tarso I, vista lateral; 6, metatarso y tarso II, vista lateral; 7, metatarso y tarso III, vista lateral; 8, metatarso y tarso IV, vista lateral; 9-10, hembra de Pirehueico; 9, region coxoesternal (detalle); 10, pedipalpo derecho, vista lateral. Diagnosis y descripcion. — El material tipico de esta especie se encuentra descolorido y en mal estado de conservation, especialmente el paratipo hembra. Los dibujos que se ofrecen en este trabajo corresponden a ejemplares de Pirehueico (MACN 8399 y 8400). Medidas en milimetros del holotipo macho en MAURY— DESCRIPCION DE NAHUELONYX AND VALDiVIONYX 75 Tabla 1. — Tarsitos. Variabilidad en numero en pata II. Numero Frequencia Nahuelonyx nasutus Vaidivionyx crassipes Machos Hembras Machos Hembras 6 0 0 0 1 7 1 2 2 7 8 9 32 16 11 9 46 55 7 3 10 28 21 1 0 11 3 0 0 0 Tabla 2. — Medidas en milimetros. Nahuelonyx nasutus Vaidivionyx crassipes Holotipo macho Hembra Holotipo macho Alotipo hembra Longitud total 4.22 4.93 4.22 4.99 Prosoma, longitud 1.41 1.28 1.60 1.60 ancho 1.79 1.86 1.92 1.98 Escudo, longitud 1.79 2.24 1.79 1.98 ancho 2.37 3.01 3.01 3.20 Pedipalpo, longitud 2.87 3.07 3.78 3.97 Trocanter 0.32 0.32 0.45 0.45 Femur 0.70 0.77 0.96 0.96 Patela 0.51 0.51 0.58 0.58 Tibia 0.64 0.70 0.83 0.83 Tarso 0.70 0.77 0.96 1.15 Pata I, longitud 5.76 5.70 6.59 6.59 Trocanter 0.45 0.51 0.51 0.58 Femur 1.22 1.28 1.54 1.54 Patela 0.70 0.70 0.83 0.83 Tibia 1.02 1.66 1.22 1.15 Metatarso 1.22 1.15 1.28 1.28 Tarso 1.15 1.02 1.22 1.22 Pata II, longitud 8.58 8.20 9.78 9.54 Trocanter 0.51 0.58 0.70 0.70 Femur 1.73 1.73 1.98 1.92 Patela 0.90 0.90 1.02 0.90 Tibia 1.41 1.34 1.60 1.54 Metatarso 1.98 1.92 2.11 2.05 Tarso 2.05 1.73 2.37 2.43 Pata III, longitud 6.08 5.96 6.85 6.78 Trocanter 0.58 0.58 0.70 0.70 Femur 1.22 1.34 1.54 1.54 Patela 0.70 0.64 0.77 0.70 Tibia 1.15 0.90 1.09 1.22 Metatarso 1.41 1.41 1.34 1.34 Tarso 1.02 1.09 1.41 1.28 Pata IV, longitud 8.55 8.96 9.79 9.31 Trocanter 0.77 0.83 0.77 0.70 Femur 1.66 1.73 2.11 1.92 Patela 0.96 0.96 0.96 0.96 Tibia 1.41 1.54 1.60 1.66 Metatarso 2.47 2.56 2.43 2.47 Tarso 1.28 1.34 1.92 1.60 Quelicero, longitud 1.86 1.98 2.37 1.92 Segmento I 0.77 0.77 0.96 0.83 Segmento II 1.09 1.21 1.41 1.09 76 THE JOURNAL OF ARACHNOLOGY la Tabla 2; la hembra medida proviene de Pirehueico (MACN 8400). La longitud total de los ejemplares estudiados vario entre 3.4 y 4.3 mm para los machos y 3.5 y 5.2 mm para las hembras; se observaron subadultos de hasta 3.4 mm de longitud. Coloracion general castano muy oscuro con jaspeado amarillento. En el prosoma se destacan, por detras y a los costados del tuberculo ocular, dos manchas ocelares amarillo intense (Fig. 1). El resto del prosoma y el escudo con manchado difuso; en algunos ejemplares hay manchas mas definidas en los hordes laterales del escudo. Tergitos libres con manchado difuso; hordes libres amarillentos. Coxas, esternon y operculo genital con fino tramado amarillento. En ambos sexos hay una mancha amarillo blancuzca en la zona de articulation del operculo genital y, exclusivamente en el macho, dos manchas paramedianas postoperculares amarillentas, (Figs. 2, 9). Queliceros y pedipalpos con fino puntillado amarillento. Trocanter, femur, patela y tibia de las patas con fino puntillado amarillento; en los metatarsos el astragalo es mas oscuro que el calcaneo, especialmente en la pata IV; en todos los metatarsos la separacion astragalo/ calcaneo marcada por una zona anular incolora (Figs. 5-8). En la pata I los tarsitos 1° y 3° oscuros y el 2° claro; en la II todos los tarsitos oscuros; en III y IV los tarsitos 1°, 3° y 4° oscuros, el 2° claro. Relacion longitud prosoma: longitud escudo entre 1:1.23 y 1:1.76. Prosoma con algunas granulaciones dispersas por detras del tuberculo ocular y un arco de tuberculos puntiagudos en todo el perimetro que se continuan, pero algo menos notables, en el escudo (Fig. 1). El tuberculo ocular, oblicuo en relacion al eje del prosoma, es conico y posee, ademas de unos pocos granulos disperses, una prominente apofisis apical. Areas del escudo marcadas por hileras transversales de gruesas granulaciones, entre las que se intercalan otras de menor tamaho. Tergitos libres con gruesas granulaciones. Esternitos y placa anal finamente granuloses (Fig. 2). Coxa I con algunos granulos ventrales pero sin tuberculos que se destaquen; coxas II a IV con unos pocos granulos ventrales; la coxa IV con tuberculos digitiformes en el horde posterior. Segmento II del quelicero liso, con una serie de pelos rigidos en la cara dorsal (Fig. 3). Pedipalpos (Figs. 4, 10) muy chicos; trocanter con dos pequenismos tuberculos dorsales y uno ventral; femur con dos o cuatro tuberculitos piliferos dorsales y otros dos o tres ventrales; patela lisa; tibia con cuatro tuberculitos piliferos en el borde ventral externo, cara ventral densamente pilosa; tarso con tres pares de pequenos tubeculos piliferos, toda la cara ventral muy pilosa. Patas (Figs. 5-8): trocanter y femur fuertemente granulosos, pero sin tuberculos mas destacados; patela y tibia algo menos granulosos y metatarso finamente granuloso, excepto el anillo de separacion astragalo/ calcaneo, que es liso. Las proporciones astragalo/ calcaneo en los metatarsos son las siguientes: pata I: 1:1.25; pata II: 1:4.50; pata III: 1:1.44 y pata IV: 1:4.71. Formula tarsal similar en los dos sexos: 3-7/11-4-4. En la Tabla 1 se ha indicado la variabilidad en el numero de tarsitos de la pata II, separado por sexo. El ovipositor (Figs. 14- 15) posee dos fuertes apofisis laterales, curvas y quitinizadas; hay cinco pares de sensilos ventrales y tres pares, algo mas debiles, de sensilos dorsoapicales. Pene (Figs. 11-13): el glande muestra la parte dorsolateral bifurcada en su extreme distal en ramas fuertemente divergentes; en la cara lateral hay una apofisis triangular de vertice dirigido hacia el extreme distal y existe un par de gruesos sensilos ubicados en el centro, paralelos y ventrales al estilo. Una profunda escotadura anterior separa la parte dorsolateral de la parte ventral; esta ultima posee una apofisis lateral de vertice dirigido hacia el extreme basal; una laminilla MAURY— DESCRIPCION DE NAHUELONYX AND VALDIVIONYX 77 Figs. 11-15. — Nahuelonyx nasutus (Ringuelet): 11-13, macho de Pirehueico; 11, glande, vista ventral; 12, glande, vista lateral; 13, glande, vista dorsal; 14-15, hembra de Pirehueico; 14, ovipositor, vista dorsal; 15, ovipositor, vista ventral. hendida longitudinalmente (la hendidura se ensancha hacia la base) y cuatro pares de sensilos: un par mayor ubicado en lateral de la laminilla y tres pares mas chicos en ventral. El estilo se presenta suavemente curvado hacia la cara ventral, con el extremo de forma compleja, semejante a un caliz. Material estudiado. — ARGENTINA: Provincia de Neuquen; Lago Queni, 2 XII 1985 (E. Maury), 1 macho (MACN 8397), Pucara, Lago Lacar, 20 km ai O de San Martin de los Andes, 19 I 1972 (L. Herman), 2 juveniles (AMNH), 4 km al O de Pucara (900 m), 21 I 1972 (L. Herman), 1 juvenil (AMNH), camino entre Pucara y Laguna Venados, 24-25 I 1972 (L. Herman), 1 hembra y 2 juveniles (AMNH), Rio Pucara, Lago Lacar, 13 I 1986 (L. Platnick, P. Goloboff y R. Schuh), 1 macho y 3 hembras (AMNH): Provincia de Rio Negro; Puerto Blest, Lago Nahuel Huapi (770 m), 2 III 1979 (Mision Cientlfica Danesa), 1 hembra (ZMC), Lago Frias, II 1950 (S. Coscaron y O. de Ferrariis), macho holotipo (MLP 24202) y hembra paratipo (MLP 24195) de Diasia nasuta Ringuelet. CHILE: Provincia de Cautln; Termas de Palquln, SE de Pucon, 17 I 1987 (E. Maury), 1 macho (MACN 8398): Flor del Lago, 15 km al NE de Viliarica, 10 XI 1985 (S. y J. Peck), 1 macho (AMNH): Provincia de Valdivia; Pirehueico, 18 I 1985 (E. Maury), 1 macho (MACN 8399), 1 XII 1985 (E. Maury), 1 macho, 4 hembras y 1 juvenil (MACN 8400): Provincia de Osorno; Los Derrumbes, 5 km al S de Termas de Puyehue, 4-5 XII 1985 (E. Maury), 2 hembras (MACN 8401), Termas de Puyehue (180 m), 24 XI 1981 (N. Platnick y R. Schuh), 4 machos, 2 hembras y 5 juveniles (AMNH), 25 XI 1981 (N. Platnick y R. Schuh), 1 macho y 4 juveniles (AMNH), 1 km al E de Termas de Puyehue (305 m), 31 I 1985 (N. Platnick y O. Francke), 2 machos y 2 hembras (AMNH), Aguas Calientes, 28 I 1986 (N. Platnick y R. Schuh), 7 machos, 8 hembras y 1 juvenil (AMNH), Antillanca (720 m), 18-24 XII 1982 (A. Newton y M. Thayer), 1 juvenil (AMNH), 20-25 XII 1982 (A. Newton y M. Thayer), 1 hembra (AMNH); 4.1 km al E de Anticura (430 m), 19-26 XII 1982 (A. Newton y M. Thayer), 1 juvenil (AMNH), 19 XII 1984 al 6 II 1985 (S. y J. Peck), 4 hembras (AMNH), 1-11 I 1986 (L. Pena), 9 machos y 8 hembras (AMNH), 19-29 X 1985 (L. Pena), 12 machos y 21 hembras (AMNH), XII 1985 (L. Pena), 15 machos y 14 hembras (AMNH), Anticura-Repucura, 6 II 1985 (S. y J. Peck), 2 hembras (AMNH), colinas al S de Maicolpue (120 m), 30 I 1985 (N. Platnick y O. Francke), 1 hembra (AMNH): Provincia de Llanquihue; 5 km al S de Ensenada, 25 I 1986 (N. Platnick y R. Schuh), 1 macho (AMNH): Provincia de Chiloe; 5 km al N de Quellon (107 m), 1 XII 1981 (N. 78 THE JOURNAL OF ARACHNOLOGY Platnick y R. Schuh), 1 macho y 1 juvenil (AMNH), Chepu (7 m), 28 XII 1981 (N. Platnick y R. Schuh), 1 hembra (AMNH), 2 II 1985 (N. Platnick y R. Schuh), 1 macho (AMNH): Provincia de Palena; Chaiten (100 m), 4 XII 1981 (N. Platnick y R. Schuh), 1 macho y 1 juvenil (AMNH), vecindades de Chaiten, 5-7 XII 1981 (N. Platnick y R. Schuh), 1 macho (AMNH), 70 km al S de Chaiten, 16 I 1986 (N. Platnick, P. Goloboff y R. Schuh), 1 juvenil (AMNH); 22 km al NE de Puerto Ramirez, 2 XII 1986 (E. Maury), 1 juvenil (MACN 8402), 28.5 km al O de Futaleufu, 16 I 1986 (N. Platnick, P. Goloboff y R. Schuh), 1 macho y 1 juvenil (AMNH). Valdivionyx , genero nuevo Diasia: Ringuelet 1959:256 (en parte, no Diasia Sorensen 1902). Especie tipo. — Valdivionyx crassipes , especie nueva, aqui designada. Etimologia. — El nombre generico Valdivionyx proviene del nombre Valdivia , como referenda al habitat de este genero, el bosque humedo valdiviano y del griego onyx: una. Distribucion. — Argentina: provincia de Rio Negro; Chile: provincias de Osorno, Llanquihue, Chiloe y Palena (Figs. 32-33). Diagnosis y descripcion. — Triaenonychinae. Triaenonychini. Prosoma de igual largo o levemente mas corto que el escudo tergal. Tuberculo ocular elevado, conico, con una corta apofisis terminal. Prosoma con una serie de tuberculos aguzados en todo el perimetro. Areas del escudo tergal poco definidas, inermes, levemente insinuadas por algunas pequeiias granulaciones. Estigmas respiratorios libres, alejados del borde posterior de la coxa IV. Femur de las patas I a IV ligeramente granuloso pero sin tuberculos que se destaquen. Operculo genital de borde libre triangular. Coxas I a IV con algunos pequenos tuberculos ventrales. Metatarsos I a IV con la separacion astragalo/calcaneo bien marcada por una zona anular deprimida e incolora; pata I con el astragalo de igual largo que el calcaneo, pata II con el astragalo menor que el calcaneo, patas III y IV con el astragalo mayor que el calcaneo. En ambos sexos, la pata I con los artejos del tarso ligeramente engrosados; en el macho las patas III y IV con los artejos del tarso muy engrosados. Distitarso de la pata I con dos segmentos; de la pata II con tres segmentos. Formula tarsal similar en los dos sexos: 3-6/10-4-4. Segmento basal de los queliceros con una apofisis terminal dorsomedial. Dimorfismo sexual: macho con los tarsos de las patas III y IV mas engrosados que en la hembra; con los pedipalpos mas robustos, especialmente femur y tibia y con el operculo genital de forma mas acusadamente triangular; hay ad e mas dos areas paramedianas postoperculares lisas y amarillentas, ausentes en la hembra. Ovipositor bilobulado, con cinco pares de sensilos ventrales y tres pares dorsoapicales. Pene con el estilo derecho; parte dorsolateral levemente bifurcada en el apice; parte ventral con una laminilla hendida longitudinalmente y con cuatro pares de sensilos. Coloracion castano muy oscura con jaspeado amarillento, una gran mancha ocelar amarilla a cada lado del prosoma. Valdivionyx se asemeja mucho en la coloracion al genero Nahuelonyx , pero una buena cantidad de caracteres morfologicos y de genitalia los separan facilmente, como puede verse en la clave adjunta. El unico genero conocido de triaenoniquido con una similar proporcion de astragalo/calcaneo en los metatarsos I a IV es Triaenonychoides H. Soares 1968. Este genero, con dos especies (Maury, en prensa, b) es exclusivo de Chile y se distingue de Valdivionyx por su mayor tamano y coloracion parduzca; ademas macho y hembra presentan MAURY— DESCRIPCION DE NAHUELONYX AND VALDIVIONYX 79 Figs. 16-20. — Valdivionyx crassipes, especie nueva: holotipo macho; 16, glande, vista ventral; 17, glande, vista lateral; 18, glande, vista dorsal; 19-20, alotipo hembra; 19, ovipositor, vista dorsal; 20, ovipositor, vista ventral. cuatro artejos tarsales en la pata I y, en el macho, los tarsos III y IV no muestran los segmentos engrosados como en Valdivionyx ; la forma del tuberculo ocular y la armadura de los pedipalpos pueden ser otros caracteres distintivos. Valdivionyx crassipes, especie nueva Figs. 16-33 Diasia nasuta : Ringuelet 1959:259 (en parte), figs. 32 c-d. Material tipico. — Holotipo macho (AMNH): Isla Tenglo, Puerto Montt, Provincia de Llanquihue, Chile; alotipo hembra (MACN 8403) y una hembra paratipo (MACN 8404): Termas de Pichicolo, Provincia de Palena, Chile; dos machos y dos hembras paratipos (AMNH) y dos machos paratipos (MACN 8405): Termas de Puyehue, Provincia de Osorno, Chile. Etimologia. — El nombre especifico crassipes proviene del latln crassus : grueso y pes: pata, haciendo referencia al engrosamiento de los tarsos III y IV en el macho de esta especie. Diagnosis y description. — Medidas en millmetros del material tipico indicadas en la Tabla 2. La longitud total de los ejemplares estudiados vario entre 4.10 y 4.86 mm para los machos y 3.52 y 5.18 mm para las hembras; se observaron subadultos de hasta 3.14 mm de longitud. Colaracion general castano muy oscuro con jaspeado amarillento. En el prosoma se destacan, por detras y a los costados del tuberculo ocular, dos manchas ocelares amarillo intenso (Fig. 21). El resto del prosoma, as! como el escudo, con manchas difusas; en algunos especlmenes hay 80 THE JOURNAL OF ARACHNOLOGY Figs. 21-31. — Valdivionyx crassipes , especie nueva: 21-28, holotipo macho; 21, cuerpo, vista lateral; 22, region coxoesternal (detalle); 23, quelicero derecho, vista lateral; 24, pedipalpo derecho, vista lateral; 25, metatarso y tarso I, vista lateral; 26, metatarso y tarso II, vista lateral; 27, metatarso y tarso III, vista lateral; 28, metatarso y tarso IV, vista lateral; 29-31, alotipo hembra; 29, region coxoesternal (detalle); 30, pedipalpo derecho, vista lateral; 31, metatarso y tarso III, vista lateral. MAURY— DESCRIPCION DE NAHUELONYX AND VALDIVIONYX 81 Fig. 32.— Ubicacion del area estudiada. Fig. 33.— Localidades estudiadas de Nahuelonyx nasutus (Ringuelet) (estrellas) y de Valdivionyx crassipes, especie nueva (circulos). 82 THE JOURNAL OF ARACHNOLOGY manchas mas definidas en los hordes laterales del escudo. Tergitos libres castano oscuro con los hordes amarillentos. Coxas y esternon con fino tramado amarillento; operculo genital con esfumado; esternitos con el horde amarillento. En ambos sexos hay una mancha amarillo blancuzca en la zona de articulation del operculo genital y, exclusivamente -en el macho, dos manchas paramediaeas postoperculares amarillentas (Figs. 22-29). Queliceros y pedipalpos con tramado amarillento. Trocanter, femur, patela y tibia de las patas con tramado amarillento; metatarsos castano esfumado; en todos los metatarsos la separation astragalo/ calcaneo marcada por un anillo de color mas claro; en el astragalo IV se nota una pseudoarticulation de color levemente mas palido (Fig. 28). Tarsitos de todas las patas con esfumado castano. Relation longitud prosoma : loegitud escudo entre 1:1 y 1:1.25. En el prosoma, detras del tuberculo ocular, hay dos hileras longitudinales de graeulaciones, levemente curvadas hacia medial; ademas de ue arco de tuberculos puntiagudos, espatiados, en todo el perimetro. El tuberculo ocular, oblicuo en relation al eje del prosoma, es conico y remata en una corta apofisis roma apical; hay tambien algunos granulitos disperses. Areas del escudo casi lisas, con algunas pequenas graeulaciones piliferas (Fig. 21). Tergitos libres, esternitos y placa anal con granulitos esparcidos (Figs. 22-29). Coxa I con algunos granules ventrales, los mayores se disponen cerca del horde anterior; coxas II a IV con algunos granules disperses; la coxa IV con tuberculos digitiformes en el horde posterior. Segmento II del quelicero casi liso, con algunos granulitos piliferos en la cara dorsal (Fig. 23). Pedipalpos (Figs. 24, 30) relativamente pequeeos, algo mas robustos en el macho, especialmeete femur y tibia. Trocanter con dos granulitos dorsales y uno ventral; femur casi liso dorsalmente, con dos o tres pequenisimos granulitos piliferos; horde ventral con tres tuberculos mas prominentes, sobre todo el basal que es aguzado; patela lisa; tibia con tres fuertes tuberculos piliferos en el horde ventral extern© y uno en el horde ventral interne; tarso con tres tuberculitos piliferos en el borde ventral externo y dos en el borde ventral interno. Patas (Figs. 22-28, 31): trocanter, femur, patela y tibia con algunas graeulaciones pero sin tuberculos que se destaquen; metatarso finamente granuloso, excepto el anillo de separation astragalo/ calcaneo, que es liso. Las proporciones astragalo xalcaneo en los metatarsos son las siguientes: pata I: 1:1; pata II: 1:5.20; pata III: 1:0.50 y pata IV: 1:0.26. Formula tarsal similar en los dos sexos: 3-6/10-4-4. En la labia 1 se ha indicado la variabilidad en el nurnero de tarsitos de la pata II, separado por sexo. Ovipositor (Figs. 19-20) bilobulado, con cinco pares de sensilos ventrales y tres pares dorsoapicales. Pene (Figs. 16-18): el glande preseeta la parte dorsolateral levemente bifurcada en el apice, terminando en dos pequenas apofisis; parte ventral con la laminilla hen <1.1 da longitudiealmeete y con cuatro pares de sensilos: un par mayor ubicado lateralmente a la laminilla y tres pares algo mas chicos ubicados ventralmente; el estilo es recto, ensanchandose levemente en el tercio distal. Material estudiado. — ARGENTINA: Provincia de Rio Negro; Lago Frias, febrero de 1950 (S. Coscaron y O. de Ferrariis), hembra alotipo (MLP 24378), 1 macho y 1 hembra paratipos (MLP s/e) de Diasia nasuta Ringeelet. CHILE: Privincia de Osorno; Termas de Puyehue (180 m), 24 XI 1981 (N. Platnick y R. Schufa), 2 machos y 2 hembras paratipos (AMNH), 2 machos paratipos (MACN 8405), 12 III 1965 (H. Levi), 2 machos y 1 hembra (MCZ), 10 km al E de Puyehue, 24 I 1951 (E. Ross y A. Michelbacher), 4 machos (CAS), Parque National Puyehue, rata a Antillanca (470 m), 20- 25 XII 1982 (A. Newton y M. Thayer), 1 hembra (AMNH), 18-24 XII 1982 (A. Newton y M. Thayer), 1 macho y 1 hembra (AMNH), Los Oerrumbes, 5 km al S de Termas de Puyehue, 4-5 XII 1985 (E. MAURY— DESCRIPCION DE NAHUELONYX AND VALDIVIONYX 83 Maury), 2 hembras y 1 juvenil (MACN 8406), Anticura, XII 1985 (L. Pena), 1 macho (AMNH): Provincia de Llanquihue; Isla Tenglo, Puerto Montt, 9 III 1962 (A. Archer), macho holotipo (AMNH): Provincia de Chiloe; Isla Chiloe, 15-18 XII 1985 (L. Pena), 1 macho (AMNH): Provincia de Palena; Termas de Pichicolo, 11 km al O de Rio Negro, 8-9 XII 1985 (E. Maury), hembra alotipo (MACN 8403), 1 hembra paratipo y 1 juvenil (MACN 8404). AGRADECIMIENTOS Este trabajo ha sido posible realizarlo gracias al envio de material por los siguient.es colegas e iestituciones: Dr. N. Platnick, American Museum of Natural History, New York (AMNH); Dr. W. Pulawski, California Academy of Sciences, San Francisco (CAS); Dr. H. Levi, Museum of Comparative Zoology, Harvard University, Cambridge (MCZ); Dr. H. Enghoff, Zoologisk Museum, Copenhagiie (ZMC) y Lie. R. Arrozpide, Museo de La Plata (MLP); otros materiales pertenecen al Museo Argentino de Ciencias Naturales, Buenos Aires (MACN). Al Sr. J. Cokendolpher, Texas Tech University, Lubbock, agradezco su intermedia- cion con el material del AMNH. LITERATURA CITADA Maury, E. A. (en prensa, a). Triaenonychidae Sudamericanos. II. El genero Diasia Sorensen 1902 (Opiliones, Laniatores). Physis. Maury, E. A. (en prensa, b). Triaenonychidae Sudamericanos. IV. El genero Triaenonychoides H. Soares 1968 (Opiliones, Laniatores). Bol. Soc. Biol. Concepcion. Maury, E. A. y A. H. Roig Alsina. 1985. Triaenonychidae Sudamericanos. I. El genero Ceratomontia Roewer 1915 (Opiliones: Laniatores). Hist. Nat., Corrientes 5(1 1):77-92. Ringuelet, R. A. 1959. Los aracnidos argentinos del orden Opiliones. Rev. Mus. Argentino Cienc. Nat., Zool., 5(2): 127-439. Soares, H. E. M. 1968. Contribuigao ao estudo dos opilioes do Chile (Opiliones: Gonyleptidae, Triaenonychidae). Pap. Avul. Zool. Sao Paulo, 21(27):259-272. Sorensen, W. E. 1902. Gonyleptiden (Opiliones Laniatores). In Ergebnisse der Hamburger Magalhaensischen Sammelreise 1892/93. II. Band., Arthropoden: pp. 1-36. Manuscript received March 1987 , revised July 1987. Jennings, D. T. and D. J. Hilburn. 1988. Spiders (Araneae) captured in Malaise traps in spruce-fir forests of west-central Maine. J. Arachnol., 16:85-94. SPIDERS (ARANEAE) CAPTURED IN MALAISE TRAPS IN SPRUCE-FIR FORESTS OF WEST-CENTRAL MAINE Daniel T. Jennings USD A, Forest Service Northeastern Forest Experiment Station USDA Building, University of Maine Orono, ME 04469 USA and Daniel J. Hilburn1 Department of Entomology Deering Hall, University of Maine Orono, ME 04469 USA ABSTRACT Spiders of 12 families, 20 genera, and 25 species were captured in modified Malaise traps deployed in spruce-fir forests of Somerset and Piscataquis Counties, Maine. Numbers of species and individuals differed between web-spinner and hunter foraging strategies. Sorensen’s similarity quotient (QS) indicated that the Malaise-trapped fauna had greater similarity to arboreal than to terrestrial spider faunas of northeastern spruce-fir forests. Spider-trap interactions include accidental capture and possibly attraction; attractive features include trap architecture, concentrated potential prey, and protective shelter. INTRODUCTION Malaise traps (Malaise 1937) and modified versions (Gressitt and Gressitt 1962; Butler 1965; Townes 1962, 1972) primarily were designed for capture of flying insects. Such traps have been acclaimed, “one of the major advances in collecting methods in this century” (Steyskal 1981, p. 225). Malaise-trap captures include numerous species of Diptera, Hymenoptera, and Lepidoptera, with lesser numbers of Heteroptera and Coleoptera (Townes 1972). Other insect orders and arthropod classes including spiders (Arachnida, Araneae) also are trapped; however, to our knowledge Malaise traps have not been purposefully used to collect spiders. During investigations of insecticidal impacts on terrestrial nontarget organisms (Hilburn 1981), modified Malaise traps were deployed in spruce-fir forests of west-central Maine. The forests were infested with the spruce budworm, Present address: Department of Agriculture and Fisheries, P.O. Box 834, Hamilton HMCX, Bermuda. 86 THE JOURNAL OF ARACHNOLOGY Fig. 1. — Modified Malaise trap for capturing insects and spiders in spruce-fir forests, west-central Maine. Choristoneura fumiferana (Clemens), the most destructive defoliator of conifers in the northeastern United States and Canada (Kucera and Orr 1981). Numerous insects and fewer spiders were captured in the Malaise traps; insect captures were summarized by Hilburn (1981). In this paper we describe the Malaise-trapped spiders, compare the trapped fauna with terrestrial- and arboreal-spider faunas of northeastern forests, and identify possible spider-Malaise trap interactions. METHODS Spiders were collected in 12 Malaise traps deployed at 12 sampling sites (1 trap/ site) in spruce-fir forests of west-central Maine near Moosehead Lake. Three sites were in Somerset County; nine were in Piscataquis County. For details of sampling sites and sampling protocol, see Hilburn (1981) and Hilburn and Jennings (1988). The Malaise traps were modifications of Townes’ (1972) design and were placed on the ground in the herb-shrub layer (Fig. 1). Spiders and insects were captured in 1-pint (0.47-liter) jars containing 70% ethyl alcohol as a killing-preservative agent. Captured specimens were sorted and identified in the laboratory. Although there were six 48-h sampling periods for each site, collected spiders were combined from all sites and over all sampling dates (21 May to 29 June 1980). Spiders were identified by the senior author; species determinations follow Kaston (1981) and other consulted sources including: Opell and Beatty (1976) for the Hahniidae; Leech (1972) for the Amaurobiidae; Chamberlin and Gertsch (1958) for the Dictynidae; Dondale and Redner (1982) for the Clubionidae; and Dondale and Redner (1978) for the Philodromidae and Thomisidae. Sexually JENNINGS AND HILBURN— SPIDERS IN MALAISE TRAPS 87 mature spiders were identified to species; most juveniles, including penultimate males, were identified to generic level However, juveniles of Philodromus spp. were assigned to species group based on carapace and leg markings (Dondale and Redner 1978); a single juvenile of R placidus Banks was identified to species based on characteristic leg markings. Chi-square analysis (x2, P - 0.05) was used to test the null hypothesis of equal spider distribution (individuals) between foraging strategies (web spinner, hunter). Because web spinners generally are more sedentary and less mobile than hunters, we suspected that Malaise-trap catches may be biased toward capture of hunters. Sorensen’s similarity quotient (QS), as defined by Price (1975, p. 341), was used to compare habitat associations (terrestrial, arboreal) for the Malaise-trapped spiders. The formula used was: QS = 2C x 100/ (A + B), where A is the number of species observed in this study, B is the number of species in the compared study (e.g., Loughton et al 1963), and C is the number of species common to both studies. Because sampling methods and intensities varied considerably among studies, the calculated QS values give only general indications of faunal affinities, not absolute associations. The comparisons were limited to spider-faunal studies of northeastern forests having spruce (Pice a spp.) and fir (Abies balsamea (L.) Mill.) components. RESULTS AND DISCUSSION Spider taxa. — Spiders of 12 families, 20 genera, and at least 25 species were collected in Malaise traps deployed in spruce-fir forests of west-central Maine (Table 1). Species composition differed by foraging strategy; species of web spinners were more prevalent (56.0% of total species) than species of hunters (44.0%). Species richness per family ranged from one (Hahniidae, Dictynidae, Thomisidae) to four (Salticidae). Spider numbers. — Of 86 total specimens collected, most (46.5%) were males; juveniles (32.6%) and females (20.9%) comprised the remainder. Four penultimate males were included in the juvenile category. The abundance of males is probably the result of greater male sexual-cursorial activity (Muma and Muma 1949); male spiders may move considerable distances in search of females. Individuals were distributed unevenly by foraging strategy, i.e., more hunters (54.6% of total specimens) were caught in the Malaise traps than web spinners (45.4%). However, the uneven distribution of individuals was not statistically significant (x2 = 0.74, df = 1, P> 0.05) between foraging strategies. Males and females of the sac spider Clubiona canadensis Emerton were by far the most commonly collected spider in the Malaise traps; this species accounted for 25.6% of all specimens. Habitat associations. — Most of the species of spiders taken in Malaise traps have been collected in other northeastern spruce-fir forests (Loughton et al. 1963; Renault 1968). Comparisons with previous spider-faunal studies indicated that the Malaise-trapped fauna had greater similarity (i.e., higher QS values) to the arboreal fauna than to the terrestrial fauna of northeastern forests (Table 2). And, by definition (QS < 50; Price 1975) the Malaise-trapped fauna was distinct from all compared terrestrial and arboreal faunas. The relatively low similarity (QS = 11.5) between pitfall collections (Hilburn and Jennings 1988) and Malaise-trap 88 THE JOURNAL OF ARACHNOLOGY Table 1. — Species and numbers of spiders (Araneae) in Malaise traps, spruce-fir forests of west- central Maine, 1980. FAMILY SPECIES AND NUMBER HAHNIIDAE WEB SPINNERS Antistea brunnea (Emerton) 1 female AMAUROBIIDAE Amaurobius borealis Emerton 1 male DICTYNIDAE Callobius bennetti (Blackwall) 3 males Dictyna phylax Gertsch & Ivie 1 male THERIDIIDAE Theridion differens Emerton 2 males LINYPHIIDAE Theridion pictum (Walckenaer) 1 female Theridion spp. 1 penult, male, 3 juv. Front inelia pyramitela (Walckenaer) 1 juv. ERIGOMIDAE Microlinypkia mandibulata (Emerton) 1 female Dismodicus bifrons decemoculatus (Emerton) 2 males, 4 females ARANEIDAE Hypsetistes florens (O.P.-Cambridge) 1 male, 2 females Undet. spp. 1 penult, male, 3 juv. Araneus sp. 1 juv. TETRAGNATHIDAE Araniella displicata (Hentz) 1 male Nuctenea sp. 1 juv. Tetragnatha versicolor Walckenaer 2 females CLUBIOMIOAE Tetragnatha sp. 1 penult, male, 5 juv. HUNTERS Clubiona canadensis Emerton 17 males, 5 females PHILODROMIDAE Clubiona kastoni Gertsch 1 male Clubiona trivialis C. L. Koch 1 male, 1 female Clubiona spp. 1 penult, male, 6 juv. Philodromus exilis Banks 3 males THOMI5IDAE Philodromus placidus Banks 1 juv. Philodromus spp. (rufus group) 2 juv. Tibellus oblongus (Walckenaer) 3 males Misumena vatia (Clerck) 2 males SALTICIDAE Eris sp. 1 juv. Meiaphidippus flavipedes (G. & E. Peckham) 1 male Meiaphidippus protervus (Walckenaer) 1 female Sitticus finschii (L. Koch) 1 male collections at the same study sites support this conclusion. The Malaise-trap spiders probably are representative of the intermediate herb-shrub layer; however, comparative studies are lacking for these strata. Some of the Malaise-trapped species are commonly associated with terrestrial habitats; others are commonly associated with arboreal habitats. The amauro- biids, Amaurobius borealis Emerton and Callobius bennetti (Blackwall), frequently are found on or near the ground (Kaston 1981); C. bennetti also occurs under loose bark of spruce and fir trees killed by the spruce budworm (Jennings, unpubl. data), and on foliage of balsam fir (Loughtoe et al 1963). Likewise, Dictyna phylax Gertsch & I vie, Araniella displicata (Hentz), Philodromus exilis Banks, P. placidus , and Meiaphidippus flavipedes (G. & E. Peckham) are most commonly found on foliage of conifers (Renault 1968; Do nd ale and Redner 1978; Jennings and Collies 1987b); rarely are these species found on the ground. A related species of Philodromus, P lutulentus Gertsch, has been taken in Malaise traps elsewhere (Dondale and Redeer 1978). JENNINGS AND HILBURN— SPIDERS IN MALAISE TRAPS 89 o 3 a E/3 a "cl T3 d KS 15 «« '? hj 3 o cG « f— ! da DO m * os ? go OS — < OS SO _ — . ^ 2 I - " • O c «a fc U © sq sd 3 £ 2 <* rn Os £ m M | .a 2 £ « £® -5. G £ P 4) S=s > ffi ~ o £ 'C « .© cs Z > C ^ O 0 M ' c S © "o S F c 1 >S ^ -o .ts « c 2 3 «s 5 > £» > 5« JT a s M P W "St «o OO SO « d 31 £ ^ .2 3 5 a o « u 3 -c § * | os « P .d q* ffl Crf d Q W U, s_ 0) 2 P Odd o -0 t. 3 ^ ^ 3 T-, O. ^ ^ a o o a C« p S Cfl W5 p « .a a a - S Pd PU DO C/3 Ph b W) a a eS a & £ £ £ £ £ ^ tp W< hi u X) X) £> P p a a o o o 0M ft* eS rt _ _ «««W0JOOO HHFPP<<< 90 THE JOURNAL OF ARACHNOLOGY A few of the Malaise-trapped spiders in west-central Maine are known to frequent both terrestrial and arboreal habitats or intermediate herb-shrub strata. The sac spider Clubiona canadensis Emerton has been taken in pitfall traps, under stones, and in leaf litter (Dondale and Redner 1982); this species also occurs on foliage of red spruce, Picea rubens Sarg. (Jennings and Collins 1987a), and balsam fir (Loughton et al. 1963; Renault 1968). Specimens of C. trivialis are common inhabitants of spruce, fir, and pine (Pinus) foliage, but also are found under loose bark, under stones, and in leaf litter. Another species of sac spider, Trachelas tranquillus (Hentz), has been taken “in the folds and crevices of Malaise traps” (Dondale and Redner 1982, p. 126); Platnick and Shadab (1974) also report this species from Malaise traps. The crab spider Misumena vatia (Clerck) has been collected commonly on flowers and foliage of many herbs, shrubs, and deciduous trees (Dondale and Redner 1978), and coniferous trees (Jennings and Collins 1987b). Tibellus oblongus is usually found in tall grass (Dondale and Redner 1978), and occasionally in pitfall traps (Varty and Carter 1974; Jennings et al. 1988). Species of Hahniidae are small spiders that spin delicate sheet webs near the ground (Kaston 1981). Antistea brunnea (Emerton) has been taken by pitfall traps in spruce-fir forests of Maine (Jennings et al. 1988; Hilburn and Jennings 1988), but not from the tree canopy layer. The erigonids also are small spiders that live chiefly under dead leaves near the ground (Kaston 1981); however, some species, e.g., Hypselistes florens (O. R-Cambridge), are taken in large numbers by sweeping bushes and grasses (Kaston 1981), and on foliage of balsam fir (Loughton et al. 1963; Renault 1968). Both species of comb-footed spiders captured in the Malaise traps, Theridion differens Emerton and T. pictum (Walckenaer), are common inhabitants of conifers (Renault 1968; Kaston 1981); however, T, differens also occurs in grass and low bushes (Kaston 1981). The bowl and doily spider, Frontinella pyramitela (Walckenaer), spins a characteristic sheet web in low branches, bushes, and tall grass (Kaston 1981), and on foliage of balsam fir (Loughton et al. 1963; Renault 1968). The platform spider, Microlinyphia mandibulata (Emerton), spins a platform-like sheet web, “usually in grass two to six inches from the ground,” (Kaston 1981, p. 124); however, this species also has been taken on foliage of balsam fir (Renault 1968). Notably absent from the Malaise-trap collections in west-central Maine were species of Agelenidae, Gnaphosidae, and Anyphaenidae. The agelenid funnel- weavers are frequently taken in pitfall traps (Carter and Brown 1973; Jennings et al. 1988; Hilburn and Jennings 1988) and on coniferous-tree foliage (Loughton et al. 1963; Renault 1968) in northeastern forests. Species of Coelotes and Wadotes are found under loose bark and stones; species of Agelenopsis spin their funnel webs in grasses and on bushes (Kaston 1981). Gnaphosid spiders also are frequently taken in pitfall traps in northeastern spruce-fir forests; species of Haplodrassus and Zelotes have been taken on foliage of balsam fir (Loughton et al. 1963; Renault 1968). Gnaphosid species reported taken in Malaise traps include: Drassodes saccatus (Emerton) (Platnick and Shadab 1976); Herpyllus ecclesiasticus Hentz (Platnick and Shadab 1977); Nodocion floridanus (Banks) (Platnick and Shadab 1980a); and Cesonia bilineata (Hentz) (Platnick and Shadab 1980b). Sergiolus cyaneiventris Simon has been collected “in insect flight traps,” (Platnick and Shadab 1981). Of these gnaphosid species, only H. ecclesiasticus has been recorded from Maine. JENNINGS AND HILBURN— SPIDERS IN MALAISE TRAPS 91 Spiders of the family Anyphaenidae are long-legged active hunters (Dondale and Redner 1982); some inhabit foliage of trees and shrubs, others are found in leaf litter and in crevices under logs and stones on the forest floor. Interestingly, none have been taken in pitfall traps or in foliage samples from northeastern spruce-fir forests. Species of anyphaenids recorded taken in Malaise traps include: Aysha gracilis (Hentz), Wulfila saltabundus (Hentz), Anyphaena pectorosa L. Koch, and Anyphaena aperta (Banks), all reported by Dondale and Redner (1982); Anyphaena maculata (Banks), Anyphaena pectorosa , Anyphaena fraterna (Banks), and Wulfila alba (Hentz), all reported by Platnick (1974). Only W. saltabundus has been taken in Maine and Nova Scotia; the other species have more southern distributions (Platnick 1974; Dondale and Redner 1982). Spider-trap interactions. — Are spiders attracted to Malaise traps? Or, is their presence in these traps accidental? Based on numbers and species collected during this study, we suggest that the interactions of spiders with Malaise traps may be more than accidental. Although some initial encounters with Malaise traps may be accidental, we suspect that spiders respond favorably to attractive features of a suitable habitat. However, this hypothesis needs testing under controlled conditions. Three possible features that may influence attraction of spiders to Malaise traps are: (1) the physical architecture of the traps, (2) the presence of abundant potential prey, and (3) the sheltered protection from the elements and from natural enemies. Spiders respond to structural features within habitats (Greenquist and Rovner 1976) and many species colonize man-made structures (Fowler 1980; Robinson 1981; Streit and Roser-Hosch 1982; Stevenson and Dindal 1981) and man-made environments (Duffey 1975). There is increasing evidence that structural features within habitats play important roles in habitat selection by spiders (Riechert and Gillespie 1986). Spiders also respond to increases in prey density (Riechert and Gillespie 1986), and abundance of prey influences habitat selection (Turnbull 1964; Riechert and Luczak 1982). Because Malaise traps attract numerous flying insects, especially Diptera, Hymenoptera, and Lepidoptera (Townes 1962), these traps are sources of aggregated prey density. Spiders respond to aggregations of prey (MacKay 1982; Riechert 1976). We suspect that spiders may be attracted to concentrations of insects, especially near the apex and catchment jar of Malaise traps. For Malaise-trap maintenance, Martin (1977, p. 27) advises, “look inside the trap, especially the entrance to the killing bottle, for spider webs, which must be removed and the spiders captured and killed, if possible.” In Maine, Hilburn observed spider webbing near the apex of a Malaise trap without catchment jar; the web was positioned to capture insects exiting from the trap. Avoidance of predators also affects habitat selection by spiders (Riechert and Gillespie 1986). In this respect, Malaise traps may provide spiders temporary shelter and protection from their natural enemies, such as birds and predatory wasps (Pompilidae and Sphecidae). The traps also may provide shelter from rain and extremes of temperature. CONCLUSION Although Malaise traps are widely used for capture of flying insects (see Steyskal (1981) for a bibliography on Malaise traps), rarely are spiders recorded 92 THE JOURNAL OF ARACHNOLOGY among the captured fauna. We found only two previous studies (Wilkinson et ah 1980: Hauge and Midtgaard 1986) that included spiders among Malaise-trap captures. A few isolated records of individual species taken in Malaise traps are found in the araneological literature (e.g., Dondale and Redner 1982; Platnick 1974; Platnick and Shadab 1976, and others); most records concern species of Anyphaenidae, Clubionidae, and Gnaphosidae. We suspect that spiders may occur more commonly in Malaise-trap collections than is reported in the entomoiogical-araneological literature. The sparsity of published information on Malaise-trapped spiders may be due to the failure of investigators to collect, identify, and report such captures. Personal communications with investigators who frequently use Malaise traps support these conclusions; both Robert W. Matthews and Richard FI Roberts have observed spiders in their Malaise traps on numerous occasions, but the spiders were not collected and identified. Finally, Malaise traps may supplement (but not supplant) other methods used for collecting and sampling spiders, especially for species in the herb-shrub layer. Malaise traps also may be useful for testing hypotheses concerning aggregation responses of spiders to increased prey densities. ACKNOWLEDGMENTS We thank Daniel Starr, USDA, National Agricultural Library, Beltsville, MD, for AGRICOLA and DIALOG literature searches. Janet J. Melvin, USDA, Northeastern Forest Experiment Station, Orono, ME, provided word processor service. Charles D. Dondale and James H. Redner, Biosystematics Research Centre, Ottawa, identified Dismodicus and Siiticus species, for which we are grateful. We also thank Robert W. Matthews, Department of Entomology, University of Georgia, Athens, GA; and Richard H Roberts (retired), USDA, Agricultural Research Service, Gainesville, FL, for permission to cite their unpublished observations. Lastly, we thank our reviewers John B. Dimond, Charles D. Dondale, G. B. Edwards, Richard R. Mason, Robert W. Matthews, William B Peck, and Richard H. Roberts for their constructive comments on an earlier draft. LITERATURE CITED Butler, G. D„, Jr. 1965, A modified Malaise insect trap. Pan-Pacific EntomoL, 41:51-53. Carter, N. E. and N. R. Brown. 1973. Seasonal abundance of certain soil arthropods in a feeitrothion- treated red spruce stand. Canadian EntomoL, 105:1065-1073. Chamberlin, R. V. and W. J. Gertsch. 1958. The spider family Dictynidae in America north of Mexico. Amer. Mus. Nat. Hist. Bull. 116. 152 pp. Dondale, C. D. and J. H. Redner. 1978. The crab spiders of Canada and Alaska (Araneae: Philodromidae and Thomisidae). Canadian Dept. Agric, PubL, 1663. 255 pp. Dondale, C. D. and J. H. Redner. 1982. The sac spiders of Canada and Alaska (Araneae: Clubionidae and Anyphaenidae). Canadian Dept. Agric. PubL, 1724. 194 pp. Duffey, E. 1975. Habitat selection by spiders in man-made environments. Proc. 6th. Inti. Arachn, Cong., Amsterdam IV. 1974:53-67. Fowler, H. G. 1980. “Trap nests” for studying desert web-building spiders (Araneae: Pholcidae). Entomol. News, 91:136-138. Freitag, R., G. W. Ozburn and R. E. Leech. 1969. The effects of sumithioe and phosphamidon on populations of five carabid beetles and the spider Trochosa terricola in northwestern Ontario and including a list of collected species of carabid beetles and spiders. Canadian Entomol, 101:1328- 1333. JENNINGS AND HILBURN— SPIDERS IN MALAISE TRAPS 93 Greenquist, E. A. and J. S. Rovner. 1976. Lycosid spiders on artificial foliage: Stratum choice, orientation preferences, and prey wrapping. Psyche, 83:196-209. Gressitt, J. L. and M. K. Gressitt. 1962. An improved Malaise trap. Pacific Insects, 4:87-90. Hauge, E. and F. Midtgaard. 1986. Spiders (Araneae) in Malaise traps from two islands in the Oslofjord, Norway. Fauna norv. Ser. B, 33:98-102. Hilburn, D. J. 1981. The effects of aerial spraying for spruce bud worm with carbaryl and Bacillus thuringiensis of non-target terrestrial arthropods. MS Thesis, Univ. of Maine at Orono. 58 pp. Hilburn, D. J. and D. T. Jennings. 1988. Terricolous spiders (Araneae) of insecticide-treated spruce-fir forests in west-central Maine. Great Lakes Entomologist, (in press). Jennings, D. T., M. W. Houseweart, C. D. Dondale and J. H. Redner. 1988. Spiders (Araneae) associated with strip-clearcut and dense spruce-fir forests of Main. J. Arachnol., 16:55-70. Jennings, D. T. and J. A. Collins. 1987a. Spiders on red spruce foliage in northern Maine. J. Arachnol., 14:303-314. Jennings, D. T. and J. A. Collins. 1987b. Coniferous-habitat associations of spiders (Araneae) on red spruce foliage. J. Arachnol., 14:315-326. Kaston, B. J. 1981. Spiders of Connecticut. 2nd. ed. Bull. Connecticut State Geol. Nat. Hist. Surv., 70. 1020 pp. Kucera, D. R. and P. W. Orr. 1981. Spruce budworm in the eastern United States. U.S. Dept. Agric., For. Serv., For. Insect Dis. Leafl. 160. 7 pp. Leech, R. 1972. A revision of the nearctic Amaurobiidae (Arachnida: Araneida). Mem. Entomol. Soc. Canada, 84. 182 pp. Loughton, B. G., C. Derry and A. S. West. 1963. Spiders and the spruce budworm. Pp. 249-268, In The Dynamics of Epidemic Spruce Budworm Populations. (R. F. Morris, ed.). Mem. Entomol. Soc. Canada, 31. 332 pp. Mac Kay, W. P. 1982. The effect of predation of western widow spiders (Araneae: Theridiidae) on harvester ants (Hymenoptera: Formicidae). Oecologia (Berlin), 53:406-41 1. Malaise, R. 1937. A new insect-trap. Entomol. Tidskr., 58:148-160. Martin, J. E. H. 1977. The insects and arachnids of Canada. Part 1. Collecting, preparing, and preserving insects, mites, and spiders. Canadian Dept. Agric. Publ., 1643. 182 pp. Muma, M. H. and K. E. Muma. 1949. Studies on a population of prairie spiders. Ecology, 30:485- 503. Opell, B. D. and J. A. Beatty. 1976. The nearctic Hahniidae (Arachnida: Araneae). Bull. Mus. Comp. Zook, 147(9):393-433. Platnick, N. 1974. The spider family Anyphaenidae in America north of Mexico. Bull. Mus. Comp. Zook, 146(4):205-266. Platnick, N. I. and M. U. Shadab. 1974. A revision of the tranquillus and speciosus groups of the spider genus Trachelas (Araneae, Clubionidae) in North and Central America. Amererican Mus. Novitates, 2553:1-34. Platnick, N. I. and M. U. Shadab. 1976. A revision of the spider genera Drassodes and Tivodrassus (Araneae, Gnaphosidae) in North America. American Mus. Novitates, 2593:1-29. Platnick, N. I. and M. U. Shadab. 1977. A revision of the spider genera Herpyllus and Scotophaeus (Araneae, Gnaphosidae) in North America. Bulk American Mus. Nat. Hist. 159:1-44. Platnick, N. I. and M. U. Shadab. 1980a. A revision of the North American spider genera Nodocion, Litopyllus, and Synaphosus (Araneae, Gnaphosidae). American Mus. Novitates, 2691:1-26. Platnick, N. I. and M. U. Shadab. 1980b. A revision of the spider genus Cesonia (Araneae, Gnaphosidae). Bulk American Mus. Nat. Hist., 165:335-386. Platnick, N. I. and M. U. Shadab. 1981. A revision of the spider genus Sergiolus (Araneae, Gnaphosidae). American Mus. Novitates, 2717:1-41. Price, P. W. 1975. Insect Ecology. John Wiley & Sons, New York. 514 pp. Renault, T. R. 1968. An illustrated key to arboreal spiders (Araneida) in the fir-spruce forests of New Brunswick. Canada Dept. Fisheries and Forestry, For. Res. Lab., Fredericton, New Brunswick. Internal. Rept. M-39, 41 pp. Riechert, S. E. 1976. Web-site selection in the desert spider Agelenopsis aperta. Oikos, 27:311-315. Riechert, S. E. and R. G. Gillespie. 1986. Habitat choice and utilization in web-building spiders. Pp. 23-48, In Spiders: Webs, Behavior, and Evolution. (W. A. Shear, ed.). Stanford Univ. Press, Stanford, California. 492 pp. Riechert, S. E. and J. Luczak. 1982. Spider foraging: behavioral responses to prey. Pp. 353-385, In Spider Communication: Mechanisms and Ecological Significance. (P. N. Witt and J. S. Rovner, eds.). Princeton Univ. Press, Princeton, New Jersey. 440 pp. 94 THE JOURNAL OF ARACHNOLOGY Robinson, J. V. 1981. The effect of architectural variation in habitat on a spider community: an experimental field study. Ecology, 62:73-80. Stevenson, B. G. and D. L. Dindal. 1981. A litter box method for the study of litter arthropods. J. Georgia Entomol. Soc., 16:151-156. Steyskal, G. C. 1981. A bibliography of the Malaise trap. Proc. Entomol. Soc. Washington, 83:225- 229. Streit, B. and S. Roser-Hosch. 1982. Experimental compost cylinders as insular habitats: colonisation by microarthropod groups. Rev. Suisse Zoo!., 89:891-902. Townes, H. 1962. Design for a Malaise trap. Proc. Entomol. Soc. Washington, 64:253-262. Townes, H. 1972. A light-weight Malaise trap. Entomol. News, 83:239-247. Turnbull, A. L. 1964. The search for prey by a web-building spider Achaearanea tepidariorum (C. L. Koch) (Araneae, Theridiidae). Canadian Entomol., 96:568-579. Varty, I. W. and N. E. Carter. 1974. Inventory of litter arthropods and airborne insects in fir-spruce stands treated with insecticides. Canadian For. Serv., Maritimes For. Res. Centr. Info. Rep. M-X- 48. 32 pp. Wilkinson, J. D., G. T. Schmidt and K. D. Biever. 1980. Comparative efficiency of sticky and water traps for sampling beneficial arthropods in red clover and the attraction of clover head catepillar adults to anisyl acetone. J. Georgia Entomol. Soc., 15:124-131. Manuscript received August 1987, no revisions. Roach, S. H. 1988. Reproductive periods of Phidippus species (Araneae, Salticidae) in South Carolina. J. Arachnol., 16:95-101. REPRODUCTIVE PERIODS OF PHIDIPPUS SPECIES (ARANEAE, SALTICIDAE) IN SOUTH CAROLINA1 Steven H. Roach Cotton Production Research Unit, Agricultural Research Service U.S. Department of Agriculture, P.O. Box 2131 Florence, South Carolina 29503 USA ABSTRACT Observations were made on the reproductive periods of nine species of Phidippus occurring in South Carolina. Phidippus audax (Hentz) oviposited from May through the following early spring, while most other species had much shorter and seasonally defined oviposition periods. The reproductive periods noted in these studies were similar to published reports from other regions where these species occur. INTRODUCTION The occurrence of Phidippus spp. in South Carolina has been reported by Roach and Edwards (1984) and Gaddy and Morse (1985). Many of these species occur in the same habitats and are often difficult to separate taxonomically. The use of genetic product analyses to identify and to study the phylogenetic relationships of many of the species that occur in the southeastern United States was reported by Terranova and Roach (1987a, 1987b). Because the immatures of several species may compete for habitat and prey concurrently, it is important to know the seasonal phenology of each species. Most studies of Phidippus spp. reproduction cycles in the literature are limited to observations on the egg sacs of individuals or a limited number of species in a geographical region. These studies were reviewed by Edwards (1980) who also reported his observations on the reproductive cycles and egg masses of Phidippus spp. occurring in Florida. In this report, I present a summary of seven years of observations on the reproductive periods of nine of the eleven Phidippus spp. occurring in South Carolina and compare these periods to those reported from other parts of each species range. METHODS During the course of collecting salticids for other studies (Roach 1983; Roach & Edwards 1984; Terranova & Roach 1987a, b), numerous specimens of Phidippus spp. were captured and held for observation. Collection methods 'in cooperation with the South Carolina Agricultural Experiment Station. This article reports the results of research only. Mention of a proprietary product does not constitute endorsement or a recommendation for its use by the US DA. 96 THE JOURNAL OF ARACHNOLOGY varied according to habitat being sampled, but were primarily by sweep-net sampling and visual searching. Spiders thus collected were placed in clear plastic containers (8 X 8 or 8 X 4 cm) and held in a programmed environmental cabinet at 27 ± 2°C, RH 50 ± 10%, and a photoperiod of 14:10 (L:D). Spiders were fed Heliothis spp. larvae approximately the same size as the spider every 2 to 3 days until the natural death of the spiders. Observations on egg sac production included in this report are from gravid females collected in the wild. RESULTS The most commonly collected species of Phidippus in field habitats in South Carolina is P. audax (Hentz). This species began egg sac formation and oviposition in early May, and continued through most of the year (Table 1). Multiple egg sacs by P. audax were common, with an average of 2.75 per female. Table 2 shows the number of eggs per sac and the relative periods of their occurrence. In these observations, the mean number of eggs per sac (60) was about equal for all except possibly the sixth, even though the range (15-164) of eggs per sac was quite variable. Phidippus clams Keyserling was most frequently collected from old field habitats and lakeshore areas, and generally shared the same habitats as P. audax. However, the seasonal reproductive cycle of P. clans was more restricted than P. audax (Table 1). P clarus females oviposted during August and September and spiderlings dispersed from September through January. Another species that occupied habitats similar to P. clams and P audax was P. princeps (Peckham & Peckham). However, it was only found in old field habitats, particularly in wooded areas, and on young pines in reforested areas. This species was common, but less generally distributed than P. clams over the areas sampled. Only two gravid females were observed in this study. During the period from February to April, one produced a single egg sac and the other produced two egg sacs. The average number of eggs per sac (32) was considerably less than that of the previous two species. Phidippus mystaceus (Hentz) has been collected only from the western foothills of South Carolina. Two gravid females included in this study were collected with egg sacs on 28 February and 23 March in Pickens County, SC by J. Brushwein. Each female produced only one egg sac, one with 76 and the other with 92 eggs (Table 1). Spiderlings from both egg sacs dispersed during mid April. Collection records from Pickens County indicated this species is found on shrubs, trees, and on the ground under protective coverings. Phidippus otiosus (Hentz) was collected state-wide and is primarily an arboreal species. This species matured during the fall and produced egg sacs from December to February (Table 1). Phidippus whitmani Peckham & Peckham was also collected statewide exclusively from woods litter, primarily in older, mixed hardwood areas. Of six females observed, oviposition occurred in July and August, and no female produced more than one egg sac (Table 1). The remaining species included in this report (Table 1) were rarely collected and thus observations on these species are limited. Phidippus putnami (Peckham & Peckham) adults were collected from low limbs on the edge of mixed ROACH— REPRODUCTIVE PERIODS OF PHIDIPPUS 97 5* S c X S a, 3 .1 8 «- > A ) >-« |L V .23 jo T3 O &p rs i..- « 1> & T3 W3 ’3 >* O .'S o 1 3 3 * s a ?! < c o r; "o s .g il « ■> °* S-> H3 4) 4> 3 > 6 « 3 23 ■ fl >> i > £***& V & 5pJ2222*-,Svo e H, b _ 5 ^ C/3 ^ ^ ^ 2 «n u. V. § ^ -. S * o i i i i i 'if ro U1 t"" *— ' VO os — < Ov —i ts r-~ — < *— > en © O © # rn # mrj^-JhoamO oomooo'^rro^f^ t-- p-- CN —I «S & «ri u 2 C/3 o* ,2 < 2 60 X X - " « Pu 60 fc . 3 B* c < < :-; V v; ^ cd S < - VO M N n — — ’ vO «N1 g S3 II « > !-=< c « 4) 60 ,0 X) o O 09 1 4) to o . Z r» \o ^ '. a. oo On D s ^ ^ o « cn QQ \o rT A v> m a — . 4) —i C/3 '. - S3 < 60 sR|s < A 3§ «A _ -a etf to 60 « « 60 ;..0 co « 60 60 5 .2 .3 QJ O* W) « i 1^ OA •5.3 co Da O <*-* « ° 3 M Q s >z ed ROACH— REPRODUCTIVE PERIODS OF PHIDIPPUS 99 hardwood areas during late summer and early fall Only one gravid female was collected with an egg sac and the spiderlings were possibly dispersing when found. This female did not produce another egg sac before dying in December. Phidippus regius C. L. Koch was collected only in coastal areas, and again only one gravid female was observed. It was collected in November and produced one egg sac in February (Table 1). Phidippus cardinaiis (Hentz) was collected only in the foothills area of the state by I. Brashweie. He collected two females with egg sacs in March and April, 1986, but unfortunately did not count the number of eggs per sac. DISCUSSION Phidippus audax occurs widely over most of the United States and several reports of seasonal occurrence are available. Gibson (1947), in Tennessee, reported that P. audax overwintered as immatures and adults but did not deposit eggs until July. Kaston (1981) indicated that adults matured in late April to early May in Connecticut and laid eggs in June and July, with single females constructing up to three egg sacs. Snetsinger (1955), in Illinois, reported that P, audax mated in May and June and deposited eggs in June and July. Edwards (1980) indicated a maturation period primarily in May and June for Florida. Taylor & Peck (1975) compared southern Texas and northern Missouri forms of P. audax , and indicated a spring maturation period with up to six egg sacs per female, averaging 41.7 to 85.5 young per egg sac. They also found that later egg sacs for each female contained fewer young than earlier deposited egg sacs. These results are similar to those found in the present study except that egg sacs deposited successively did not contain fewer eggs than those produced earlier, with the possible exception of the sixth egg sac. Oviposltlon, by R dams was observed in August-September In South Carolina. Kaston (1981), in Connecticut, reported mating of P dams in June and observation of an egg sac in late July; he also indicated a P. dams female was collected on 31 August while guarding an egg sac with 47 eggs. Snetsinger (1955) observed mating of P darns in Illinois from late June to early August and egg sac formation August to October. Edwards (1980) reported P dams matured in July and August In Florida. Although there is variability in maturation in this species, egg sac formation occurs primarily from late July to October in the geographic area from Florida to Illinois. Kaston (1981) reported that P. princeps matured in April and May in Connecticut and laid eggs as early as May; he also reported collecting a female guarding eggs on 10 June. Cutler (1965), In New York, reported seeing adults in September. In Florida, P, princeps Is uncommon, but Edwards (1980) stated that the oviposltlon period was from May to July. All of these periods are somewhat later than the February-April oviposition period noted in South Carolina. Berry (1970) reported collecting a single adult of P mystaceus in June In the Piedmont region of North Carolina. Kelley (1979) collected several mature specimens in Pickens County, SC, and indicated the breeding season is from April to May in that area. Edwards (1980) indicated the oviposition period of R mystaceus in northern Florida is October through May. Kaston (1981) Indicated that this species is rare in Connecticut. Present observations on spiders collected 100 THE JOURNAL OF ARACHNOLOGY in Pickens County, SC indicated that oviposition by P. mystaceus occurs in February and March. Thus, in its eastern range, P. mystaceus apparently matures during the fall and winter and breeds during spring and early summer. Phidippus otiosus (also known as P pulcher) is primarily a Southeastern species and the only extensive phenology of this species reported in the literature is by Edwards (1980) for northern Florida. He indicated this species matures from September to November and oviposits from January to June. In South Carolina, this species matured in the fall and the collected females oviposited from December through February. The only information found in the literature on the reproductive cycle of P putnami is for northern Florida (Edwards 1980). He reported that this species matured in July and August, and oviposited from August through October. The collection of adults in late summer and early fall, along with the collection of an egg sac in October, indicate a similar cycle in South Carolina. Phidippus regius is primarily a southeastern species which matures during September and October and oviposits from October to June in Florida (Edwards 1980). My observations indicate a similar cycle for P. regius in the coastal area of South Carolina. Phidippus whitmani is a widely distributed species that matures in May in Connecticut (Kaston 1981), June in North Carolina (Berry 1970), and May or June in Florida (Edwards 1980). In South Carolina, the pattern is similar, with adults collected during the summer and oviposition observed during July and August. Kaston (1981) reported that P cardinalis adults were collected in Connecticut from late May to October, while Edwards (1980) indicated that in Florida the species matures from September to November and oviposits February through May. Phidippus cardinalis was only collected with egg sacs during March and April in South Carolina and apparently is not a common species over most of the state. Thus, this species may have a somewhat later maturity period in its more northern range. Two other Phidippus species, P. apacheanus Chamberlin & Gertsch and P. purpuratus Keyserling, also occur in South Carolina but no adult females have been collected and observed so their phenology in the region is unknown. However, Edwards (1980) indicated that P. apacheanus matured in September- October in northern Florida, while Gardner (1965) reported that the species matured during the same period in the area around Reno, Nevada. Phidippus purpuratus is more common in the northeastern states and adults occur from May-September in Connecticut (Kaston 1981). In summary, the reproductive periods of the various Phidippus species vary in time of occurrence and possibly in other phenological parameters. The information found in the report should be useful in predicting what stage of each species will be present in various habitats during certain periods of the year. ACKNOWLEDGMENTS The assistance of G. B. Edwards, Florida State Collection of Arthropods, in confirming the taxonomy of species included in this report is gratefully acknowledged. Special thanks is extended to J. Brushwein, Clemson University, for supplying several of the spiders collected in Pickens Co., South Carolina. ROACH— REPRODUCTIVE PERIODS OF PHIDIPPUS 101 LITERATURE CITED Berry, J. W. 1970. Spiders of the North Carolina piedmont old field communities. J. Elisha Mitchell Sci. Soc., 86:97-105. Cutler, B. 1965. The jumping spiders of New York City. J. New York Entomol. Soc., 73:138-143. Edwards, G. B. 1980. Taxonomy, ethology, and ecology of Phidippus (Araneae: Salticidae) in eastern North America. Ph.D. dissertation, Univ. of Florida, Gainesville. Gaddy, L. L. and J. C. Morse. 1985. Common spiders of South Carolina with an annotated checklist. South Carolina Agric. Experiment Station, Tech. Bull. 1094, 182 pp. Gardner, B. T. 1965. Observations on three species of Phidippus jumping spiders (Araneae: Salticidae). Psyche, 72:133-147. Gibson, W. W, 1947. An ecological study of the spiders of a river terrace forest in western Tennessee. Ohio J. Sci., 47:38-44. Kaston, B. J. 1981. Spiders of Connecticut. Connecticut State Geol. & Natural Hist. Bull., 70:1-874. Revised 1981. Kelley, R. W. 1979. Niche partitioning among spiders of a granite outcrop. M.S. Thesis, Clemson Univ., Clemson. Roach, S. H. and G. B. Edwards. 1984. An annotated list of South Carolina Salticidae (Araneae). Peckhamia, 2:49-57. Roach, S. H. 1983. Susceptibility of Oxyopes salticus Hentz and Phidippus audax Hentz (Araneae: Oxyopidae, Salticidae) to fenvalerate, azinphosmethyl, and methyl parathion. J. Georgia Entomol. Soc., 18:323-326. Snetsinger, R. 1955. Observations on two species of Phidippus. Entomol. News, 66:9-15. Taylor, B. B. and W. B. Peck. 1975. A comparison of northern and southern forms of Phidippus audax (Hentz). J. Arachnol., 2:89-99. Terranova, A. C. and S. H. Roach. 1987a. Electrophoretic key for distinguishing South Carolina species of the genus Phidippus (Araneae: Salticidae) as spiderlings and adults. Ann. Entomol. Soc. America, 80:346-352. Terranova, A. C. and S. H. Roach. 1987b. Genetic differentiation in the genus Phidippus (Araneae, Salticidae). J. Arachnol., 15:385-391. Manuscript received August 1987, no revision. Bennett, R. G. 1988. The spider genus Cybaeota (Araneae, Agelenidae). J. Arachnoh, 16:103-119. THE SPIDER GENUS CYBAEOTA (ARANEAE, AGELENIDAE) Robert G. Bennett Department of Environmental Biology University of Guelph Guelph Ontario Canada NIG 2W1 ABSTRACT Cybaeota Chamberlin and Ivie, 1933 (a genus of small, Nearctic, woodland spiders) is revised to include four species: Cybaeota caiearata, the type species, was described by Emerton in 1911, and C nana, C. munda , and C. shastae were described by Chamberlin and Ivie in 1937. Cybaeota concolor Chamberlin and Ivie, 1937 is synonymized under C. nana . Cybaeota vancouverana and C. wasatchensis (both of Chamberlin and Ivie, 1937) are synonymized under C. shastae. The relationship of Cybaeota to other Cybaeinae is discussed. INTRODUCTION In 1911 j H. Emerton described what is now the type species of Cybaeota as Liocranum calcaratum . He placed it in the Ciubionidae, apparently because of the similarity of this cryptic, eastern North American, species to certain clubionid spiders (e.g., Scotinella Banks) in size and the possession of conspicuous pairs of ventral macrosetae on various leg segments. Some years later R. V. Chamberlin and W. Ivie (1933), citing the presence of an unpaired third tarsal daw and the general similarity of the male palpus to that of Cybaeus L Koch, transferred this species to the Agelenidae and placed it in a new genus Cybaeota. Since 1933 Cybaeota. usually has been considered a member of the subfamily Cybaeinae (currently considered to encompass the genera Cybaeina , Cybaeota , and Cybaeozyga of Chamberlin and Ivie, and Cybaeus ) of the Agelenidae (Roewer 1954; Bonnet 1956; but see Lehtinen 1967 and Brignoli 1983). Subsequently Chamberlin and Ivie (1937) described C. concolor , C. munda , C nana , C shastae , C. vancouverana and C. wasatchensis from about a dozen specimens collected from British Columbia, California, and Utah. These species were diagnosed on the basis of abdominal pigmentation variations and small genitalic differences. As is often found in the taxonomic work of Chamberlin (in particular) and Ivie the descriptions are terse and vague and the drawings difficult to interpret for specimen identification. The present paper is the first in a series planned to sort out the general tangle of cybaeiee systematics and test the hypothesis of cybaeine monophyly. Cybaeota is a distinct grouping and is probably monophyletic. Genitalic apomorphies of the genus are: (1) the structure of the retrolateral tibial apophyses and the position of the bristly seta between them (Fig. 17), and (2) the structure and placement of the spermathecae and connecting ducts (Figs. 25, 28, 38). 104 THE JOURNAL OF ARACHNOLOGY The putative monophyly of the taxon Cybaeinae including Cybaeota is less well-supported. The similarities in the general structure of the male palpus shared by Cybaeota and Cybaeus are also seen in other, more distantly related genera such as Alt el la and Devade (both of Simon) in the Dictynidae, or Cicurina Menge and Tegenarla Latreille in the Agelenidae. Cybaeota strongly resembles Cybaeina in the arrangement of the ventral tibial and metatarsal macrosetae (Fig. 9) and in the fine structure of their sockets (Fig. 10), but these characters are also seen in various clubionid genera (e.g., Scotinella as mentioned above) and some other divergent agelenids (e.g., Ethobuella Chamberlin and Ivie and Cicurina ) as well as in Liocranoides Keyserling (Tengellidae) and Ischnothyreus Simon (Oonopidae). These characters are probably present in combination in other genera as well. Conspicuous, paired, ventral tibial macrosetae are of widespread but scattered distribution amongst spiders and the socket reinforcements are present in all genera possessing such macrosetae of which I had specimens to study (i.e., those listed above). The intriguing distribution of these characters suggests that they are homoplasies (or perhaps shared piesiomorphies) and probably are not indicators of close relationship. Although no good synapom or- pines can be found to support the inclusion of Cybaeota in the Cybaeinae, neither have any been found which demonstrate a closer relationship of Cybaeota to any other taxa. Cybaeota is therefore left in the Cybaeinae. In this revision three new synonyms are proposed, reducing the number of recognized species of Cybaeota from “seven species described and several others known” (Roth and Brame 1972) to four. Cybaeota concolor is synonymized under C nano; and C. wasatchensis and C. vancouverana under C shastae. The collecting activities of V. D. Roth and W. J. Gertsch have been largely responsible for boosting the number of Cybaeota specimens available for study. Because of this increase, pigmentation differences used by Chamberlin and Ivie to delimit various species can be seen to be clinal variations within species. This revision has resulted from the study of about 350 specimens from my personal collection (RGB) or kindly lent by the following institutions and individuals; the American Museum of Natural History (AMNH), Dr, N. L Platnick; the California Academy of Sciences (CAS), Dr. W. J. Pulawski; the Canadian National Collection of Insects, Arachnids, and Nematodes (CNC), Dr. C. D. Do o dale: the Museum of Comparative Zoology (MCZ), Dr. H. W. Levi; Dr. Robin E. Leech (REL); and Mr. Vincent D. Roth (V DR). Methods, — Specimens were examined and measured under a stereo dissecting microscope with an ocular micrometer reticle attached. Measurements are accurate to 0.01 mm. Identifications were made through the examination of male and female genitalia (dissected from the spiders and cleared in clove oil) under a compound microscope. The small size of these spiders makes identification with a dissecting microscope difficult. Drawings were made either with the aid of a drawing tube attached to the compound microscope or a squared grid reticle in one eyepiece of the dissecting microscope. Scanning electron micrographs were made with a Hitachi S-570 SEM. Abbreviations used in text are as follows; CL, CW (carapace length and width); SL, SW (sternum length and width). Other abbreviations are explained in figure legends. Standard postal abbreviations are used for states and provinces. Statistics are presented as sample range (mean ± standard deviation). Measurements are in millimeters. BENNETT— REVISION OF CYBAEOTA 105 Genus Cyhaeota Chamberlin and Ivie Liocranum (in part): Emerton, 1911:402, Plate V, figs. 4, 4a-f. Cybaeota Chamberlin and Ivie, 1933:3, figs. 1-10, type species Liocranum calcaratum Emerton, 1911, by monotypy; Chamberlin and Ivie, 1937:226, figs. 68-84; Roewer, 1954:87; Bonnet, 1956:1298; Lehtinen, 1967:226; Roth and Brame, 1972:17, figs. 5, 23-24; Brignoli, 1983:483; Roth and Brown, 1986:3. Diagnosis. — Male with characteristic distal and medial retrolateral tibial apophyses, with single bristly seta located between them (Fig. 17); female with simple genitalia, copulatory opening single, with two, short connecting ducts each leading to single, large, circular, heavily sclerotized spermatheca (Figs. 25, 28, 38). Description. — Small spiders, with carapace lengths averaging 0.74 (male) to 1.08 mm (female); females usually slightly larger than males. Carapace (Figs. 1, 3) darkly pigmented around eyes, pale yellow elsewhere, longer than wide, glabrous except for small number of setae along midline and around eyes; dorsal groove short, longitudinal. Usually eight eyes (Figs. 1, 3) in two rows (one specimen of C. shastae known with posterior median eyes missing); posterior row longer than anterior; both rows slightly recurved in dorsal view; in frontal view anterior row straight, posterior row recurved; anterior median eyes reduced; anterior laterals largest; posteriors subequal, somewhat smaller than anterior laterals; median ocular quadrangle widest posteriorly, about twice height of clypeus. Promargin of cheliceral fang furrow with three subequal teeth (Fig. 3), retromargin with two to five small teeth. Sternum (Fig. 2) shield-shaped, extending posteriorly short distance between coxae IV, nearly as wide as long, pale yellow, lightly clothed with fine setae. Labium (Fig. 2) short, wider than long. Serrula (Fig. 16) well developed. Legs pale yellow, without markings; I and IV longest, subequal, III shortest; I and II conspicuously setose, femur I with two (occasionally 1) distal prolateral macrosetae, other macrosetae ventral, tibia I 2-2-2-2-2, metatarsus I 2-2-2, tibia II 2-2-2-2-1, metatarsus II 2-2-1, tibia III 1-2-1; all tibial and metatarsal macrosetal sockets reinforced (as in Fig. 10). Each tarsus and metatarsus usually with two trichobothria dorsally (Fig. 13), arranged as in typical agelenids, with distal one longer than proximal. Trichobothrial sockets and tarsal organs typically araneomorph (Figs. 14, 15). Abdomen (Figs. 1, 2) rounded, unornamented, concolorous to strongly patterned (variable within species, Figs. 6-8), lightly clothed with fine setae; spiracle (Figs. 2, 11) just anterior to and as wide as colulus, which is represented by two setae; anterior spinnerets (Figs. 2, 12) broad, separated by about width of colulus, as long as posterior spinnerets; posterior spinnerets narrow, separated by width of anal tubercle; median spinnerets small, contiguous; apical segments of all spinnerets subequal, much shorter than basal segments. Epigynum simple, externally marked by transverse (Fig. 25) or inverted “U- shaped” (Fig. 27) copulatory opening; shape and position of spermathecae and connecting ducts usually discernible through integument (Figs. 2, 30-34); bursa a shallow pocket (Fig. 25) or nearly absent; connecting ducts short, sinuous (most noticeably in anterior or posterior view), separately joined to anterior margin of bursa (Fig. 25) or to anterolateral (Figs. 36, 37) or posterolateral (Fig. 27) margins of copulatory opening when bursa reduced; spermathecae simple, large, 106 THE JOURNAL OF ARACHNOLOGY rounded, heavily sclerotized, contiguous (Figs. 25, 26) or moderately separated; single fertilization duct exiting each spermatheca posteriorly (Fig. 27). Male pedipalp (see Fig. 4 for view of expanded palpal organ) simple, without patellar apophyses, with distal and medial retrolateral tibial apophyses uniform among species (Fig. 17); basal haematodocha well-developed (with petiole apparently incorporated onto proximal surface), merging with narrow ringlike subtegulum; subtegulum connected to broad, rounded tegulum by inconspicuous middle haematodocha; embolus short, stout, continuous with surface of tegulum (Fig. 19) (Gering [1953] incorrectly described the embolus of Cybaeota as terminating in a long filament such as in Wadotes Chamberlin; see Bennett 1987); conductor (Figs. 21, 22) flexibly attached (by distal haematodocha?) to surface of tegulum, with broad, shallowly excavated plate dorsal to tip of embolus, with two arms, prolateral arm varying according to species, retrolateral arm dagger-shaped; receptaculum seminis (Fig. 5) visible ih palpi cleared in clove oil, well-sclerotized throughout, coiled through —540°, fundus “s-shaped”, lying deep within subtegulum, reservoir in close association with outer margin of tegulum through —360°, ejaculatory duct “s-shaped” at base of embolus, opening just proximal to embolus tip. Natural history notes. — The orientation of palpal sclerites on the partially expanded palpus of one male C. nana (Fig. 18) suggests a functional relationship between the retrolateral arm of the conductor and the medial retrolateral tibial apophysis. During inflation of the basal haematodocha the conductor is forced proximally along the tibia until the retrolateral arm of the conductor and the medial retrolateral tibial apophysis engage. This action should impart some amount of rigidity to the cymbium as the embolus is inserted into the epigynum. The cryptic nature of all species of Cybaeota is probably responsible for their rare appearance in collections. However, within particular microhabitats, species of this genus may be dominant members of the arthropod community as has been demonstrated for other “rare” organisms (e.g., see Bennett 1985 and discussion under C. shastae). The tiny spiders of this genus are found in leaf litter, moss on tree trunks, and other debris on the floor of Nearctic forests. The species are concentrated in western North America from Utah west to California and coastally north to Alaska (Figs. 40-42). One species occurs in the northeastern United States and adjacent regions of Canada (Fig. 39). KEY TO SPECIES OF CYBAEOTA (Male of C. munda unknown) 1. Prolateral arm of conductor bifid (Fig. 19). Spermathecae large, nearly contiguous (Figs. 25, 26). NE. USA and adjacent areas of ON and PQ (Fig. 39) calcarata Prolateral arm of conductor not bifid. Spermathecae smaller, separated by about one-half their diameter. W. North America 2 2. Prolateral arm of conductor pointed and directed towards retrolateral arm (Figs. 21, 22). Connecting ducts joining copulatory opening posterolaterally (Figs. 27-29). AK to S. CA with (apparently) disjunct population in UT (Figs. 41, 42) shastae BENNETT— REVISION OF CYBAEOTA 107 Prolateral arm of conductor otherwise. Connecting ducts joining copulatory opening anterolaterally. Not known north of S. BC (Fig. 40). 3 3. Prolateral arm of conductor knob-like and directed ventrally (Figs. 23, 24). Connecting ducts not extending well into spermathecae in ventral view (Figs. 37, 38). Relatively small species (avg. female carapace length 0.8 mm). S. BC to S. CA and N. UT (Fig. 40). .nana Male unknown. Connecting ducts extending well into spermathecae in ventral view (Figs. 35, 36). Relatively large species (avg. female carapace length 1.1 mm). S. OR and mid-coastal CA (Fig. 40). munda Cybaeota calcarata (Emerton) Figs. 19, 20, 25, 26, 30, 39 Liocranum calcaratum Emerton, 1911:402, Plate V, figs. 4, 4a-f. C. calcarata : Chamberlin and Ivie 1933:4, figs. 1-10; Roewer 1954:87; Bonnet 1956:1298; Roth and Brown 1986:3. C. calcaratum : Kaston 1976:37, figs. 31-32. Diagnosis. — Male with bifid tip on prolateral arm of conductor (Fig. 19). Female with relatively large, nearly contiguous spermathecae (Figs. 25-26). Description. — As for genus. Male: N~1 including male syntype. CL 0.92-1.13 (0.99+0.07), CW 0.77-0.88(0.80+0.04), SL 0.60-0.70(0.63+0.04), SW 0.57-0.62 (0.60+0.02). Syntype CL 1.13, CW 0.88, SL 0.70, SW 0.62. Retrolateral arm of conductor with ventral longitudinal keel. Female : N- 20 including female syntype. CL 0.96-1.13 (1.04+0.04), CW 0.73- 0.90 (0.83+0.04), SL 0.59-0.70 (0.66+0.03), SW 0.56-0.65 (0.61+0.02). Syntype CL 1.13, CW 0.87, SL 0.70, SW 0.62. Distribution and natural history. — Cybaeota calcarata is the only species in this genus known from eastern North America (Fig. 39). It has been collected from forest floor litter and moss in widely scattered locales in Ontario, southern Quebec, Newfoundland, northern Michigan (Chickering 1935), New York, New Hampshire, and Massachusetts (Kaston 1948). Collection evidence suggests a year-round presence of both sexes with mature males being common only in the summer. Material examined. — Type series : two syntypes, NEW HAMPSHIRE; Coos Co., Great Gulf, Mt. Washington, 1 VIII 1910 (J. H. Emerton), 1 male, 1 female (MCZ). Mote : Following Coddington (1986:4) I prefer, in this case, not to designate a lectotype and paralectotype from the syntypes. CANADA: NF; Baie Verte Jet., 14 VIII 1984 (L. Hollett), 1 male (CNC), E of Daniels Hbr , 16 VIII 1984 (L. Hollett), 1 male, 1 female (CNC) 7 km S Pasadena, 49'00"N/57'36"W, 17 VIII 1984 (L. Hollett), 1 female (CNC), Little Barachois Brook, 20 VIII 1984 (L. Hollett), 2 females (CNC), Caribou Lk., 48'38"N/55'01"W, 24 IX 1984 (L. Hollett), 1 male (CNC), Crabbes R., 48'13"N/58'52"W, 14 VIII 1985 (L Hollett), 2 males, 1 female (CNC). ON; Algoma , Batehawana, 29 VII 1948 (W. Gertsch, W. Ivie, T. B. Kurata), 1 female (AMNH); Nipissing, Sproule Bay, Lk. Opeongo, Algonquin Pk., 26 VI-7 VII 1945 (W. Ivie, T. B. Kurata), 1 male, 4 females (AMNH), S Tea Lk., Algonquin Pk., 3-10 VII 1945 (W. Ivie, T. B. Kurata), 1 male, 1 female (AMNH), point W of Ko-Ko-Ko Bay, Lk. Temagami, 15-25 VIII 1948 (W. J. Gertsch, W. Ivie, T. B. Kurata), 1 male, 2 females (AMNH), Lk. Opeongo, Algonquin Pk., 17 VIII 1948 (W. J. Gertsch, T. B. Kurata), 1 male, 5 females (AMNH); Ottawa/ Carleton, Kinburn, in pine duff, 8 IV 1962 (J. E. H. Martin), 1 male (CNC); Thunder Bay, 3 mi. NW Finmark, N48:34/ W89:50, 23 VII 1965 (J. and W. Ivie), 1 female (AMNH). PQ; St. Hippolyte, 25 VI 1974 (M.-C. Tarrisants), 1 female (CNC). USA: NH; Cheshire , Mt. Monadnock, 13 VI 1947 (A. M. Chickering), 1 female (MCZ). NY; Albany, Rensselaerville, Huych Preserve, 8 VII 108 THE JOURNAL OF ARACHNOLOGY Figs. 1, 2. — Cybaeota munda, female, Pinnacles Nat. Mon. CA: 1, dorsal view; 2, ventral view. Fig. 3. — Cybaeota nana, female, Josephine Co. OR, face and chelicerae, frontal view. Scale markers = 0.1 mm. 1948 (Bishop), 1 female (AMNH); Tompkins, 1 male, 1 female (AMNH). NO LOCALE; AC 3222, #1425 (Horace Britcher), 1 female (AMNH). Cybaeota nana Chamberlin and Ivie Figs. 3, 4-8, 18, 23, 24, 34, 37, 38, 40 Cybaeota concolor Chamberlin and Ivie, 1937:227, figs. 77, 78; Roewer 1954:87; Bonnet 1956:1298; Roth and Brown 1986:3. NEW SYNONYMY C. nana Chamberlin and Ivie, 1937:229, figs. 74, 75, 79, 80; Roewer 1954:87; Bonnet 1956:1298; Roth and Brown 1986:3. Diagnosis. — Male with retrolateral arm of conductor smoothly curved, ventrally directed, knob-like (Figs. 23, 24). Female with spermathecae separated by approximately one-half their diameter, and with connecting ducts joined to copulatory opening anterolaterally (Figs. 37, 38). Description. — As for genus. Male: N= 21 including holotype. CL 0.66-0.92 (0.75+0.05), CW 0.55-0.73 (0.60+0.04), SL 0.43-0.61 (0.49+0.03), SW 0.43-0.53 (0.47+0.02). Holotype CL 0.70, CW 0.55, SL 0.43, SW 0.43. BENNETT— REVISION OF CYBAEOTA 109 Figs. 4-8. — Cyhaeota nana: 4, left male palpal tarsus with partially expanded genital bulb, retrolateral view; 5, receptacuium seminis of left genital bulb, ventral view, relative positions of conductor, embolus and tegulum indicated by dotted lines; 6-8, female abdomens, dorsal views indicating pattern variation within single population, Lost Lk. ID. Scale markers = 0.05 mm. BH=basal haematodocha, Oconductor, CY=cymbium, E=embolus, ED=ejaculatory duct, F=fundus, P=petiole, R=reservoir, RC=retrolateral arm of conductor, ST=subtegulum, T=tegulum. Female: N— 44 including holotype of C. concolor. CL 0.77-0.91 (0.83+.04), CW 0.60-0.73 (0.66+0.03), SL 0.47-0.57 (0.53+0.03), SW 0.46-0.55 (0.50+0.02). Holotype of C. concolor CL 0.90, CW 0.74, SL 0.56, SW 0.53. Distribution and natural history. — This species is known from extreme SW British Columbia south to S California with scattered inland records from E Washington, W Idaho, and N Utah (Fig. 40). Cyhaeota nana appears to be absent from mid-coastal California. With the probable exception of the S California coast, C. shastae is sympatric with C. nana throughout the range of the latter. 110 THE JOURNAL OF ARACHNOLOGY Figs. 9-11. — Cybaeota shastae , male, Victoria BC: 9, left tibia L proiateral view; 10, same, macroseta base and socket; 11, colulus setae and spiracle, ventral view. Fig. 12. — C. shastae , female, Josephine Co. OR, spinnerets, ventral view. CS=colulus setae. Specimens are usually taken from forest floor litter. At Corvallis, Oregon three females were found in a wood rat nest. Both sexes have been collected year-round but mature males are rarely collected in the first half of the year. Notes on synonymy. — Cybaeota concolor has page precedence over C. nana but, if retained, the former name could lead to the erroneous supposition that this species is concolorous. Chamberlin and Ivie (1937) named C nana for a pair of spiders which they perceived as abdominal coloration variants of C. shastae . It is virtually impossible BENNETT— REVISION OF CYBAEOTA 111 Figs. 13-15. — Cybaeota shastae, female, Josephine Co. OR, tarsus IV: 13, dorsal view; 14, bothrium and hair; 15, tarsal organ. Fig. 16. — C. shastae , male, Victoria BC, serrula, right palpal endite, ventral view. TO=tarsal organ, TR=trichobothrial base. non-arbitrarily to assign specimens of Cybaeota to any particular species on the basis of abdominal pattern and coloration. The genitalia of C. nana and C. concolor are identical and examination of all specimens with “nana/concolor”- like genitalia has shown a wide range of abdominal patterns. Groups of specimens from single collection locales (e.g., Cedar Lake, Stevens Co., WA; Lost Lake, ID; and City Creek, Salt Lake Co., UT) show great variability (Figs. 6-8), in one case from virtually concolorous to heavily patterned. There is a clinal trend observable across the range of this species: concolorous abdomens are 112 THE JOURNAL OF ARACHNOLOGY Fig. 17. — Cybaeota shastae, male, Josephine Co. OR, left palpal tibia, retrolateral view. Fig. 18. — C. nana, male, Los Angeles Co. CA, left palpal tibia and genital bulb, retrolateral view, showing interlocking of retrolateral arm of conductor with medial retrolateral tibial apophysis. DT, MT=distal and medial retrolateral tibial apophyses, PC=prolateral arm of conductor. prevalent in the eastern part of the range (Utah), to the west abdominal patterns become more distinct and common as the coast is approached. (There is also an east-west clinal gradation in size: larger individuals are generally eastern — Utah females average CL 0.87 mm, coastal females average CL 0.81 mm.) The conformity of genitalia of specimens previously assigned to C. nana and C. concolor combined with the clinical variability in abdominal pigmentation justifies the synonymy of C. concolor under C. nana. Material examined. — Types : Holotype of C. nana , BRITISH COLUMBIA; west side of Saanich Inlet, near Victoria, 14 IX 1935 (R. V. Chamberlin and W. Ivie), 1 male (and 1 allotype female) (AMNH). Holotype of C. concolor , UTAH; Salt Lake Co., Mill Creek Canyon, Wasatch Mtns., near Salt Lake City, no date (R. V. Chamberlin), 1 female (AMNH). USA: CA; Humboldt, Carlotta, 15 IX 1961 (W. Ivie, W. Gertsch), 1 male, 1 female (AMNH); Los Angeles, Los Angeles Nat. For., 22 VI 1957 (I. Newell), 2 males, 3 females, 2 imm. (AMNH), 6 VII 1957 (I. Newell), 3 males, 3 females (AMNH); Nevada, Sardine Valley, 14 mi. NNE Truckee (A. Grigarick), 2 females, 1 imm. male (AMNH); Riverside, San Jacinto Mtns., VII 1952 (R. X. Schick), 1 female (AMNH); Shasta , Burney Falls, 18 VI 1954 (E. Schuster), 2 males, 2 females (AMNH); Tulare, 10 mi. W Johnsondale, 15 IX 1959 (W. Gertsch, V. Roth), 2 males, 1 female (AMNH); Ventura , summit Mt. Pinos, W of Lebec, 15 IX 1959 (W. Gertsch, V. Roth), 1 male, 6 females (AMNH). ID; Lost Lk., 27 VII 1939 (W. Ivie), 2 males, 5 females, 2 imm. (AMNH); Adams, Evergreen Camp, upper Weiser R., 17 X 1944, 4 females (AMNH). NE; Washoe, Hwy 27, 19 mi. SW Tahoe Jctn., 8420', 15 VIII 1968 (R. E. and A. V. Leech), 1 male (REL). OR; Benton, N of Corvallis, McDonald For., 3 XI 1949 (V. D. Roth), 1 female, 1 imm. (CAS), W of Corvallis, 44'33"N/ 123'22"W, 20 III 1937 (J. C. Chamberlin), 1 male, 1 female (AMNH), Corvallis, 24 IV 1949 (V. D. Roth), 1 female (CAS), 26 XI 1950 (V. D. Roth), 2 males, 6 females, 3 imm. (CAS), 21 V 1952 (Roth, Birge), 3 females (CAS), 9 mi. W Philomath, 29 VII 1953 (W. J. and J. W. Gertsch), 2 females (AMNH); Josephine, summit of Wolf Ck. Rd„ 42'38"N/ 123'23"W, 12 V 1947 (I. M. Newell), 1 female (AMNH); Marion , Marion, 24 IV 1941 (J. C. Chamberlin), 1 male, 2 females (AMNH); Washington , Hillsboro, N45:30/W122:58, 1937 (J. C. Chamberlin), 2 females (AMNH). UT; Daggett, Rt. 44, 38 mi. N Vernal, 7200', 2 VIII 1959 (C. C. Hoff), 1 female (AMNH); Salt Lake, 3 mi. up City Ck. Cn., 40'47"N/ BENNETT— REVISION OF CYBAEOTA 113 Figs. 19, 20. — Cyhaeota calcorata , male syntype, Coos Co. NH, genital bulb: 19, ventral view including cymbium; 20, retrolateral view. Figs. 21, 22. — C. shastae, holotype male, Siskiyou Co. CA, conductor and embolus: 21, ventral view; 22, retrolateral view. Figs. 23, 24. — C. nana , male: 23, holotype, Saanich Inlet BC, conductor and embolus, ventral view; 24, “Redwoods” CA, genital bulb, retrolateral view. Scale markers=0.05 mm. 111'48'W, 25 VI 1962 (W. Me), 21 males, 8 females (AMNH), Mill Ck. Cm, 40'40*N/ 1 1 F45"W, 1910- 1925 (R. V. Chamberlin), 1 female (AMNH), 25 V 1924 (R. V. Chamberlin), 1 female (AMNH); Utah , Timpanogos PL, American Fork Cm, 19 VIII 1941 (J. C. Chamberlin, W. Ivie), 1 female (AMNH). WA; Kitsap , N48/W123, 1 male (AMNH); Pierce , Tacoma, 9 VIII 1929 (R. V. Chamberlin), 1 female 1 14 THE JOURNAL OF ARACHNOLOGY Figs. 25, 26. — Cyhaeota calcarata, cleared epigyna: 25, ventral view; 26, St. Hippolyte PQ, dorsal view. Figs. 27-29. — C. shastae, cleared epigyna: 27, Yosemite Nat. Pk., ventral view; 28, Weed CA, slightly posterior of ventral view; 29, same, dorsal view. Scale markers=0.05 mm. B=bursa, CD=connecting duct, CO=copulatory opening, EG=epigastric groove, FD=fertilization duct, S=spermatheca. (AMNH); Stevens, 10 IX 1963 (J. and W. Ivie), 1 female, 1 imm. (AMNH), Cedar Lk., 48'45"N/ 1 1 7'36"W (J. and W. Ivie), 4 females (AMNH), 48'55"N/ 1 17'36'W, V 1962 (W. Ivie), 1 female (AMNH), 48'56"N/117'36"W, 30 IX 1964 (J. and W. Ivie), 5 females (AMNH). Cybaeota munda Chamberlin and Ivie Figs. 1, 2, 33, 35, 36, 40 Cybaeota munda Chamberlin and Ivie, 1937:228, figs. 83, 84; Roewer 1954:87; Bonnet 1956:1298; Roth and Brown 1986:3. Diagnosis. — Male unknown. Female with connecting ducts intruding into spermathecae (Figs. 35, 36). Description. — As for genus. Female : N- 4. CL 1.01-1.17 (1.11), CW 0.81-0.92 (0.87), SL 0.62-0.75 (0.70), SW 0.59-0.65 (0.62). Holotype CL 1.13, CW 0.86, SL 0.68, SW 0.61. Distribution. — Cybaeota munda has been collected near San Francisco and from southwestern Oregon (Fig. 40). It is the only Cybaeota species known from mid-coastal California. In Oregon, C. munda is sympatric with both C. nana and C. shastae. Material examined. — Holotype : CALIFORNIA; San Mateo Co., La Honda, 1920-1921 (J. C. Chamberlin), 1 female (AMNH). BENNETT— REVISION OF CYBAEOTA 115 USA: CA; San Benito , Pinnacles Nat. Mon. (W. Gertsch, V. D. Roth), 1 female (AMNH). OR; Douglas , 5 mi. W Drain, 29 V 1948 (Roth, Brown), 1 female (CAS); Josephine , Grave Ck., 10 mi. E Placer, 22 VII 1962 (V. D. Roth), 1 female (CAS). Cybaeota shastae Chamberlin and Ivie Figs. 9-17, 21, 22, 27-29, 31, 32, 41, 42 Cybaeota shastae Chamberlin and Ivie, 1937:227, figs. 68-70; Roewer 1954:87; Bonnet 1956:1298; Roth and Brown 1986:3. C. wasatchensis Chamberlin and Ivie, 1937:227, figs. 71-73, 76; Roewer 1954:88; Bonnet 1956:1298; Roth and Brown 1986:3. NEW SYNONYMY C. vancouverana Chamberlin and Ivie, 1937:228, figs. 81, 82; Roewer 1954:87; Bonnet 1956:1298; Roth and Brown 1986:3. NEW SYNONYMY Diagnosis. — Male with pointed prolateral arm of conductor deflected towards retrolateral arm (Figs. 21, 22). Female with spermathecae separated by about one- half their diameter, connecting ducts attached to posterolateral margins of copulatory opening (Figs. 27-29). Description. — As for genus. Male : N= 24 including holotypes of C. shastae and C. wasatchensis . CL 0.81-0.96 (0.87+0.04), CW 0.62-0.78 (0.68+0.04), SL 0.53- 0.62 (0.57+0.02), SW 0.48-0.59 (0.52+0.03). Holotype CL 0.81, CW 0.64, SL 0.55, SW 0.52. Holotype of C. wasatchensis CL 0.94, CW 0.78, SL 0.61, SW 0.59. Female : 7V=61 including holotype of C. vancouverana. CL 0.86-1.09 (0.92+0.06), CW 0.65-0.87 (0.72+0.05), SL 0.55-0.73 (0.59+0.03), SW 0.51-0.65 (0.55+0.03). Holotype of C. vancouverana CL 0.87, CW 0.68, SL 0.57, SW 0.52. One male (CA; Shasta Co., Lassen Pk., 19 IX 1961) lacks posterior median eyes. Distribution and natural history. — This is the most commonly collected species of Cybaeota. It is known from scattered locales along the Alaska panhandle, on Vancouver Island (British Columbia), and the Olympic Peninsula of Washington (Fig. 42). South from Washington, this species is found along the coast and inland to N California and south through the Sierra Nevada to S California (Fig. 41). A possibly disjunct population is known from the vicinity of Salt Lake City, Utah. Cybaeota nana is sympatric with C. shastae from S Vancouver Island southwards throughout the range of the latter. Cybaeota shastae is probably common all along the British Columbia and Alaska panhandle coastlines. Berlese funnels produced good samples of this species (as well as other “rare” spiders such as Ethobuella tuonops) from moss taken from the trunks of red alder (Alnus rubra Bong.) and broadleaf maple ( Acer macrophyllum Pursh) on S Vancouver Island. Mossy red alders are common along the BC and Alaska panhandle coasts. Cybaeota shastae was the most numerous spider in these Berlese samples. Both sexes have been collected throughout the year. However, mature males are common only in late summer and fall. Mature males apparently appear earlier in more northerly parts of the species’ range. Notes on synonymy. — The three names C. shastae, C. wasatchensis , and C. vancouverana all refer to specific locales where each putative species was found. As there is no other reason to prefer one name over the others, C. shastae is 116 THE JOURNAL OF ARACHNOLOGY Figs. 30-34. — Cyhaeota , uncleared epigyna, ventral views: 30, C. calcarata, St. Hippolyte PQ; 31, C shastae with “vancouverana”-type pattern, Victoria BC; 32, C. shastae with no pattern, Weed CA; 33, holotype of C. munda with vestige of “vancouverana”-type pattern, La Honda CA; 34, C. nana, Stevens Co. WA. Figs. 35, 36. — C. munda, cleared epigyna, ventral views: 35, holotype, La Honda CA; 36, Josephine Co. OR. Figs. 37, 38. — C. nana, cleared epigyna, ventral views: 37, Lost Lk. ID; 38, Tacoma WA. Scale markers=0.05 mm. chosen as senior synonym because of its page precedence. These new synonyms are here established for the same reasons as discussed under C. nana. Variability from nearly concolorous to strongly patterned abdomens is seen in groups of specimens from Echo Summit, Eldorado Co., CA; Hughes Canyon, Salt Lake Co., Utah; and especially Shaver Lake, Fresno Co., CA. However, specimens from Alaska, British Columbia, and Washington are all strongly BENNETT— REVISION OF CYBAEOTA 117 Fig. 39. — Distribution of Cyhaeota calcarata in eastern North America (inset-Newfoundland). Hollow circles-literature records (Chickering 1935; Crosby and Bishop 1928). patterned with the spinnerets encircled with pigment and a typical “Cheshire cat face” on the epigastric area. As well, most concolorous or faintly patterned individuals come from the possibly disjunct population in Utah (originally described as C. wasatchensis). Spiders from this population are generally larger than the coastal spiders (average female CL 1.05 mm versus 0.91 mm). The lack of specimens from S Idaho and N Nevada makes a definite conclusion with respect to the clinal nature of the variability of size and abdominal pattern (as well as the disjunct nature of the Utah population) difficult. Still, I feel the observed pattern variability in other species of Cybaeota and the clinal variation in pattern and size observed in C. nana over the range it shares with C. shastae coupled with the identical morphology of the genitalia of specimens previously placed in C. wasatchensis , C. Vancouver ana, and C. shastae justifies the identity of all such species with C. shastae . Material examined. — Types : Holotype of C. shastae, CALIFORNIA; Siskiyou Co., Weed, 8 IX 1935 (W. Ivie and R. V. Chamberlin), 1 male, 2 females (allotype and paratype) (AMNH). Holotype of C. wasatchensis , UTAH; Salt Lake Co., Hughes Canyon, Wasatch Mtns., 20 V 1934 (Ivie and Rasmussen), 1 male (plus 1 female allotype) (AMNH). Holotype of C. vancouverana, BRITISH COLUMBIA; Sidney, 16 IX 1935 (R. V. Chamberlin and W. Ivie), 1 female (AMNH). CANADA: BC; Vancouver Is., Bowser, 25 VI 1955, 1 female (CNC), Cowichan Lk. Exp. Stat., 25 VII 1975 (REL), 2 females, 3 imm. (CNC), Kyuquot, 50'00"N/ 127'25"W, 2 V 1952 (S. L. Neave), 1 female (CNC), 22 IV 1959, 1 female (AMNH), 19 V 1959, 2 females, 1 imm. male (AMNH), Shawnigan Lk., 9.1 mi. W E+N RR tracks, Pt. Renfrew Rd., 14 VIII 1985 (R. G. Bennett), 4 females, 17 imm. (RGB), Sidney, 16 IX 1935 (R. V. Chamberlin, W. Ivie), 1 female (AMNH), Victoria, XI 1975 (D. State), 1 female (CNC), Victoria, Francis Regional Pk., Munn’s Rd., 2-12 VIII 1985 (R. G. Bennett), 12 males, 23 females, 13 imm. (RGB), Victoria, Coldstream Pk. (A. P. Mackie), 16 I 1975, 1 female, 14 IV 1975, 2 females, 24 IV 1975, 2 females, 23 VII 1975, 1 female, 7 VIII 1975, 1 male, 1 imm. (all CNC), 23 IX 1975 (B. Ainscough), 1 female (CNC). USA: AK; Admiralty Is., Middle Hbr., 20 VI 1932 (A. Hasselborg), 2 males (AMNH), Admiralty Is., VI 1933 (Sheppard), 1 male, 1 female (AMNH), Juneau, 28-29 IV 1945 (J. C. Chamberlin), 1 imm. (AMNH). CA; “Redwoods”, 1 male (AMNH); Eldorado, Lk. Tahoe, Echo Summit, 7382', 2 IX 1961 (W. J. Gertsch, W. Ivie), 1 male, 1 female (AMNH), Meyers, 7000', 25 VI 1953 (V. Roth), 1 male (CAS); Fresno, Shaver Lk., 12 IX 1959 (W. J. Gertsch, V. Roth), 7 males, 6 females, 1 imm. (AMNH); Humboldt, Trinidad, 16 VII 1968 (W. Ivie), 2 females (AMNH); Shasta, Lassen Vole. Nat. Pk. 7000', 19 IX 1961 (W. J. Gertsch, W. Ivie), 1 118 THE JOURNAL OF ARACHNOLOGY Fig. 40. — Distribution of Cybaeota munda (crosses) and C. nana (circles) in western North America. Figs. 41, 42. — Distribution of C shastae: 41, western USA; 42, BC and southern AK. male (AMNH), Lassen PL, 2 mi. NE Manzanita LL, 6150', 8 VIII 1968 (R. E. and A. V. Leech), 1 male (REL); Sierra, 2 mi. N Calpine, 6 IX 1959 (W. J. Gertsch, V. Roth), 2 females (AMNH); Siskiyou , Bartle, 18 IX 1961 (W. Ivie, W. J. Gertsch), 1 female (AMNH), Mt. Shasta, Panther Meadow Rd., 41'23"N/ 122T2"W, 17 IX 1961 (W. J. Gertsch, W. Ivie), 1 female (AMNH), Weed, 8 IX 1935 (R. V. Chamberlin, W. Ivie), 1 female (AMNH); Tulare , 6 mi. W Johnsondale, Double Bunk Meadows, 15 IX 1959 (V. Roth, W. J. Gertsch), 1 female (AMNH); Tuolumne , Yosemite Nat. PL, Aspen Valley, 11 VIII 1931 (W. Ivie), 2 males, 2 females (AMNH). OR: Boyer (45N/123W ?), 10 VIII 1933 (J. C. Dirks), 1 female AMNH, 15 mi. W Burnt Woods, 30 XII 1945 (R. Post), 1 female (AMNH); Benton , Corvallis, 12 V 1953 (V. Roth), 1 female (CAS), 9 mi. W Philomath, 29 VII 1953 (W. J. and J. W. Gertsch), 1 female (AMNH); C005, Bridge, Myrtlewood Camp, 27 VIM VIII 1955 (V. Roth), 1 female (CAS); Douglas , Loon LL, 1 VII 1959 (L. M. Smith), 1 female (AMNH); Jackson , 20 mi. NE Ashland, 1 IX 1959 (W. J. Gertsch, V. Roth), 1 female (AMNH); Linn , Berlin, 23 IV 1954 (Roth, Davis), 1 female (CAS), Santiam Pass, Suttle LL, 27 V 1947 (V. Roth, F. Beer), 1 female (CAS), Santiam Pass, Tombstone Prairie, 13 VIII 1949 (V. Roth), 1 female (CAS); Josephine , 1 male, 1 female, 2 imm. (AMNH); Marion , Salem, 1 V 1954 (V. Roth), 1 female (CAS). UT: Salt Lake , 40'N/lll'W, 3 males, 5 females (AMNH), Hughes Can., nr. Hoiladay, 20 V 1934 (W. Ivie), 1 male, 2 females (AMNH), Mill Ck. Can., 8 IV 1932 (W. Ivie), 1 female (AMNH), 1-2 mi. up Mill Ck. Can., 21 VIII 1941 (J. C. Chamberlin), 1 male, 1 female (AMNH). WA; Jefferson , Olympic Nat. Pk. Hoh R., 3 VIII 1954 (C. J. Goodnight), 4 females (AMNH). ACKNOWLEDGMENTS Mr. Ken Baker and Ms. Sandy Smith instructed me in SEM usage and helped in the production of SE micrographs. Mr. Doe Hamilton provided darkroom expertise. Drs. J. A. Beatty, J. P. Bogart, C. D. Dondale and S. A. Marshall and one anonymous person reviewed drafts of this paper and provided valuable BENNETT— REVISION OF CYBAEOTA 119 criticism. These folks, as well as the people and institutions who provided specimens, are all warmly thanked for their contributions to this revision. Mr. Vince Roth is particularly thanked for making his considerable cybaeine collection and notes available to me. This research has been funded in part by a Natural Sciences and Engineering Research Council of Canada post graduate scholarship. A generous Exline- Frizzell Grant from the California Academy of Sciences supported a collecting trip to British Columbia and Washington. This paper is the first part of my ongoing Ph.D. research of cybaeine agelenid systematics, supervised by Dr. S. A. Marshall. LITERATURE CITED Bennett, R. G. 1985. The natural history and taxonomy of Cicurina hryantae Exline (Araneae, Agelenidae). J. Arachnol, 13:87-96. Bennett, R. G. 1987. The systematics and natural history of Wadotes (Araneae, Agelenidae). J. Arachnol., 15:91-128. Bonnet, P. 1956. Bibliographia Araneorum. Toulouse. 2(2):9 19-1 926. Brignoli, P. M. 1983. A Catalogue of the Araneae Described Between 1940 and 1981. Manchester Univ. Press, 755 pp. Chamberlin, R. V. and W. Ivie. 1933. A new genus in the family Agelenidae. Bull. Univ. Utah, 24(5): 1 - 17. Chamberlin, R. V. and W. Ivie. 1937. New spiders of the family Agelenidae from western North America. Ann. Ent. Soc. America, 30:211-241. Chickering, A. M. 1935. Further additions to the list of Araneae from Michigan. Pap. Michigan Acad. Sci., 20:583-587. Coddington, J. A. 1986. The genera of the spider family Theridiosomatidae. Smithson. Contrib. Zool, 422:1-96. Crosby, C. R. and S. C. Bishop. 1928. Araneae. Pp. 1034-1074, In A List of the Insects of New York. (M. D. Leonard, ed.). Cornell Univ. Agr. Exper. Sta., Mem. 101. Emerton, J. H. 1911. New spiders from New England. Trans. Connecticut Acad. Arts Sci., 16:383-407. Gering, R. L. 1953. Structure and function of the genitalia in some American agelenid spiders. Smithsonian Misc. Coll., 121(4): 1-84. Kaston, B. J. 1948. Spiders of Connecticut. Connecticut St. Geol. Nat. Hist. Surv. Bull, 70:1-874. Kaston, B. J. 1976. Supplement to the Spiders of Connecticut. J. Arachnol, 4:1-72. Lehtinen, P. T. 1967. Classification of the cribellate spiders and some allied families, with notes on the evolution of the suborder Araneomorpha. Ann. Zool. Fenn., 4:199-468. Roewer, C. F. 1954. Katalog der Araneae. Bruxelles. Vol. 2a: 1-923. Roth, V. D. and P. L. Brame. 1972. Nearctic genera of the spider family Agelenidae (Arachnida, Araneida). Amer. Mus. Novit., 2505:1-52. Roth, V. D. and W. L. Brown. 1986. Catalog of Nearctic Agelenidae. Occ. Papers Mus., Texas Tech Univ., 99:1-21. Manuscript received May 1987, revised September 1987. 1988. The Journal of Arachnology 16:121 RESEARCH NOTES DRAGONFLY PREDATION UPON PHIDIPPUS AUDAX (ARANEAE, SALTICIDAE) Dragonfly adults are aerial predators capable of capturing prey in the air or from exposed surfaces. Spiders that hunt on exposed surfaces or that balloon from prominences should be potential prey for dragonflies. A review of the predators of spiders (Bristowe 1941) indicates that dragonflies are very seldom recorded capturing spiders, with only three observations listed from British Guiana, Costa Rica, and India. Reviews of the known prey of adult dragonflies (Corbet 1962, 1980), reveal only one record of predation on spiders, that of a Megalagrion sp. removing a salticid from a fern leaf in Hawaii (Williams 1936). Members of the Salticidae may be more exposed to dragonfly predation than other spiders that hunt in the canopy of the herbaceous layer, due to their general lack of crypticity as compared to the Thomisidae and to their relatively active mode of hunting. The Salticidae literature is equally depauperate in records of dragonfly predation. In a review of the ecology of Phidippus spp. in eastern North America, Edwards (1980) reports his own observation of an adult Erythemis simplicicollis (Say) (Libellulidae) preying upon an immature Phidippus pulcherrimus Keyserling. Apparently there is some risk associated with a dragonfly attempting to capture a jumping spider. Fitch (1963) observed an adult Phidippus audax (Hentz) jumping several inches into the air in unsuccessful attempts to capture adult dragonflies overhead and on other occasions observed P audax carrying dragonflies. Edwards (1980) presents two additional records of P. audax and P otiosus (Hentz) capturing adult Libellulidae. The purpose of this report is to document the behavior of a salticid in the presence of patrolling adult dragonflies and to record an instance of successful spider capture by an adult dragonfly. During the period 15-29 October 1986, visual censuses of foliage arthropods were conducted daily in a 0.1 hectare plot in Washington County, Mississippi (Young in Prep.). This plot contained a variety of weed species and three rows (length - 25 m) of nectaried cotton that had not been picked and was reflowering. Each day several individuals of Epiaeschna heros Fab. (Aeschnidae) were observed patrolling lengthwise the rows of cotton, flying 0.3-0. 7 m directly above each row. Individuals of P. audax on many occasions were also observed near the very top of these cotton plants, usually in a position that gave them some protection from the rear and that allowed them to view an adjacent plant and some of the leaf surfaces below them. Individuals of P. audax seemed to be quite capable of detecting an approaching dragonfly at a distance of approximately 3 m, perhaps aided by a moving silhouette of the dragonfly against a bright sky background. At that distance, the spider oriented its body so as to be directly 1988. The Journal of Arachnology 16:122 facing the oncoming dragonfly and assumed a position indicating a readiness to jump. Dragonflies were not observed to alter their flight path when approaching and passing over P. audax individuals, and the spiders were not observed to jump. On several occasions when a dragonfly seemed to be moving rather slowly along the row, P. audax individuals continually reoriented themselves so as to be facing the dragonfly at all times. On those warm and sunny days in which the wind exceeded approx. 3 mph, P. audax individuals frequently occurred at the top of plants in a ballooning posture. During one census of a single cotton row (25 m), 17 of 21 P. audax were at the top of plants spinning silk lines for either ballooning or for traverse lines to adjacent plants. The typical posture involved the downward inclination of the cephalothorax 40-80° below the horizontal, with the abdomen pointed upward in a near vertical position. When P. audax assumes this position an aerial predator, approaching from the spider’s rear or ventral side, might be able to avoid detection and affect capture. At 1000 hours, 19 October 1986, an adult female P. audax was in a ballooning posture on top of a seedhead of Johnson-Grass ( Sorghum halepense) at a height of 2 m. The wind was from the west and the spider was positioned such that the ventral side of its abdomen was facing west. An adult of E. heros was observed approaching the spider from the west at a height of 2 m and was first seen by us when it was 5 m from the spider. The dragonfly flew in a straight line to the spider, grabbed it in its legs, and continued flying in the same direction and at the same height until it was out of sight at a distance of approx. 75 m. The body length of the E. heros was probably in the range of 82-91 mm with a wing span > 110 mm (Needham and Westfall 1955). The body length of the P audax female was probably about 13 mm, based on an average value from 15 adult females captured and measured during the census period. Our observations on the response of P. audax to foraging dragonflies and the successful capture of P, audax by E. heros lead us to believe that dragonflies in late summer and fall may be significant predators on spiders attempting to disperse. LITERATURE CITED Bristowe, W. S. 1941. The Comity of Spiders, Vol. II. Ray Society, London. Corber, P. S. 1962. A Biology of Dragonflies. Quadrangle Books, Chicago. Corbet, P. S. 1980. Biology of Odonata. Ann. Rev. Entomol., 25:189-217. Edwards, G. B. 1980. Taxonomy, ethology, and ecology of Phidippus (Araneae: Salticidae) in eastern North America. Ph.D. thesis, Univ. Florida, Gainesville. Fitch, H. S. 1963. Spiders of the University of Kansas Natural History Reservation and Rockefeller Experimental Tract. Univ. Kansas Mus. Nat. Hist. Misc. Publ., 33:1-202. Needham, J. G. and M. J. Westfall, Jr. 1955. A Manual of the Dragonflies of North America. Univ. California Press, Berkeley. Williams, F. X. 1936. Biological studies in Hawaiian water-loving insects. Part II. Odonata or dragonflies. Proc. Hawaiian Entomol. Soc., 9:273-349. Orrey P. Young and Timothy C. Lockley, Paul’s Cove, Greenville, Mississippi 38701 USA. Manuscript received May 1987, revised August 1987. 1988. The Journal of Arachnology 16:123 NOTES ON AGGREGATIONS OF LEIOBUNUM (OPILIONES) IN THE SOUTHERN U.S.A. Aggregations of phalangids of the genus Leiobunum C. L. Koch are known from caves and mines (Holmberg et al. 1984; Mitchell and Reddell 1971), lake shores (Bishop 1950), rock overhangs (Newman 1917), buildings (McAlister 1962), and vegetation (Edgar 1971; Wagner 1954). Such formations may represent: (1) diurnal retreat aggregations (McAlister 1962; Wagner 1954), covering sometimes 25 ft2 in area (Newman 1917); (2) overwintering aggregations, for which densities of nearly three individuals/ cm2 are reported (Holmberg et al. 1984); (3) smaller and more loosely organized groups apparently associated with mating (Edgar 1971). Phalangids in retreat aggregations are quiescent except for movements associated with changes in microclimate (Holmberg et ah 1984; McAlister 1962). When disturbed, however, individuals move their bodies to-and-fro. This “bobbing” can quickly spread through an entire cluster of thousands, presumably by mechanical stimuli transmitted via legs (Newman 1917). This behavior is often accompanied by the release of volatile compounds (Holmberg et al. 1984; Ekpa et al. 1985), which may act to deter predators (Blum and Edgar 1971), although functions associated with intraspecific communication have been suggested for phalangid secretions (Bishop 1950). While observing phalangids in the southcentral U.S.A., I noted both homospecific and heterospecific aggregations of Leiobunum species. Aggregations are reported for the first time for Leiobunum flavum Banks and Leiobunum speciosum Banks. Immatures of another species, Leiobunum townsendi Weed, whose prodigious aggregating habit is described elsewhere (McAlister 1962; Mitchell and Reddell 1971), are confirmed to form aggregations. Identifications of species are based on Bishop (1949) and Davis (1934) with reference to the nomenclatural changes proposed by Cokendolpher (1984). Homospecific aggregations. — I removed an aggregation of L. townsendi from the duck blind at The University of Texas Brackenridge Field Laboratory, Austin, Travis Co., Texas at 1600 hours on 20 September 1985. This diurnal retreat aggregation was a single layer of 167 males and 157 females covering an area of 300 cm2 on the underside surface of the wooden platform. This density (1.1 individuals/ cm2) has also been calculated by Holmberg et al. (1984) for the less tightly formed clusters of Leiobunum paessleri overwintering in caves and mines of British Columbia. He also reports equal proportions of the sexes. I found late instars of L. townsendi in aggregation on the stone block restroom at the trailhead to the falls at Pedernales Falls State Park, Blanco Co., Texas. At 1200 hours on 18 April 1987, I counted 331 individuals of this species on the building, 282 within 14 aggregations (of from 6 to 70 individuals) and 49 outliers. All were in shade on the east, west, or north sides. The majority of them (64.3%) were in the window recesses, which are poorly exposed, the rest on the upper half of the walls. Two clusters of 19 and 21 were sampled and contained immatures, the ages of which were estimated morphometrically (Edgar 1971, table 3). A sample of seven individuals taken from one of these included one fourth, three sixth, and three seventh (penultimate) instars. I located two other clusters of L. townsendi along the trail to the falls. One of these contained 31 phalangids 1988. The Journal of Arachnology 16:124 loosely aggregated on the brick wall at the falls overlook. The second was more dense (like that on the duck blind), containing 150 individuals clinging to the underside of a rock ledge about 50 m upriver from the first. Both groups were composed largely of late instars. I found another cluster of this species at the boulder field above the falls. Within a shallow crevice in a large rock, 15 phalangids had formed a continuous single file. From this I sampled nine penultimates and one adult male. Another cluster on the smooth, lower face of the rock, just above the river, consisted of about 50 phalangids, at least nine of which were sixth or seventh instars. I found two small clusters of Leiobunum aldrichi Weed in the late morning of 19 July 1986 at the summit of Mount Magazine, Logan Co., Arkansas. Both males and females were loosely grouped, making contact only with the tarsi. One cluster of about 20 had formed on the underside of a rock ledge just above the outflow of Brown Springs. Another group of six was seen about a kilometer from the first in a hollowed-out area on the rock face near the head of the Cove Lake Trail. Diurnal retreat aggregations have not yet been reported for L. aldrichi and there is both field and laboratory evidence that northern populations show diurnal activity (Bishop 1950; Edgar and Yuan 1968; Fowler and Goodnight 1974). Bishop (1950) observed clusters of this species forming late in the day at the edge of a lake in a New York beech-hemlock forest. These were apparently associated with relief from unfavorable microclimate and lasted until the next morning. Edgar (1971) observed gatherings of this species (on tree trunks in a Michigan woodland) that were suggestive of prenuptial behavior. These consisted of groups of from 3 to 58 males and females which remained together for a few days until males began attempting to copulate. I kept records of activity for this species in an oak-pine forest in Tennessee from May through August, 1981. During 46 sample days, I observed five copulation attempts (between 1935 hours and 0150 hours) but none were associated with clusters. During the daytime, when L. aldrichi is inactive on trunks, I never observed more than three individuals spaced closely enough to contact legs. It is, therefore, impossible to identify this species with a single, regularly occurring type of aggregation. An aggregation of about 100 phalangids was found on the wall of a wooden shelter on the morning of 21 July 1985 near Lake Seminole, Jackson Co., Florida (L. Hribar, pers. comm.). I identified a female specimen from the aggregation as L. speciosum. (Voucher specimen is deposited at the Texas Memorial Museum). A photograph showed that this was probably a single species group but sexes could not be distinguished. Heterospecific aggregations. — At Caddo Lake State Park, Harrison Co., Texas, I found a highly clumped distribution of three species of Leiobunum under the eaves of a campground shelter. On 25 June 1986 at 2100 hours, 12 well defined aggregations were present, separated from one another by rafters. Single individuals were seen crawling along the edges of the roof. By visually comparing the area covered by each cluster to that for one I counted, I was able to estimate the numbers of phalangids within clusters. Starting at the NE corner and moving counterclockwise around the shelter, I estimated clusters of 25, 30, 250, 100, 175, 225, 300, 150, 20, 50, and 25 individuals. (The one at the SW corner was too loosely formed to estimate.) The majority of phalangids were on the north and west sides of the shelter. Density within groups was less than one phalangid/cm2. 1988. The Journal of Arachnology 16:125 A sample from the cluster of 50 showed 18 males and four females of L. flavum . (Voucher specimens are deposited at the Texas Memorial Museum). This species accounted for at least 90% of all phalangids in any given cluster. The rest were Leiohunum vitiation (seen in four clusters) and L. townsendi (seen in two). The cluster at the SW corner contained all three species. Aggregations of L. flavum have not previously been reported. Small bisexual clusters of L. vittatum were seen in summer by Edgar (1971). In checking the aggregations for activity at 2200 hours, I saw three phalangids at the perimeter of the loose cluster feeding together on an insect. I did not see copulating pairs in or near any of the groups. When I disturbed one large cluster by moving my pencil through it, a chain of phalangids dropped and hung by the legs from those above. Within seconds, the suspended ones became disentangled, fell to the ground, and began to crawl toward the shelter and up the wall. At 2300 I noted no changes in the aggregations, but at 0900 hours there was a marked difference in their dispersion. The loosely formed aggregation had disbanded, while another group had decreased in numbers from 100 to 50 individuals. Two clusters had increased in size, one from 150 to 250, the second from 20 to 125. All clusters were more dense (> one individual/ cm2) than at night and showed no activity. No solitary phalangids could be found on the shelter. I made a final check of the aggregations at 1230 hours but saw no change. I observed aggregations consisting of both males and females of L. flavum and L. vittatum at two locations in western Arkansas, each occurring beneath the eaves of campground buildings. These sightings were made at 1000 hours on 20 July 1986 at Cove Lake, Logan Co. and at 1100 hours on the following day at DeQueen Lake, Sevier Co. Conclusion. — Aggregations of Leiobunum are more common than has been reported. I here add L. flavum (in Arkansas and Texas) and L. speciosum (in Florida) to the list of Leiobunum species forming diurnal retreat aggregations. Clusters of more than one phalangid species are also reported for the first time. Aggregations were observed in spring (when immatures may be present in them) and in summer, primarily on structures providing considerable shade. All were diurnal except for a group of nocturnal clusters at one location, which were more active and loosely organized. My thanks to Dr. William Reeder and James Reddell of the Texas Memorial Museum, Austin, for supplying helpful advice and collecting materials, to Larry Hribar, Dept, of Entomology, Auburn Univ., Alabama, for the specimen and photograph from Florida, and James Cokendolpher, Dept, of Entomology, Texas Tech Univ., Lubbock, for a critical review of the manuscript. LITERATURE CITED Bishop, S. 1949. The Phalangida (Opiliones) of New York. Proc. Rochester Acad. Sci., 9:159-235. Bishop, S. 1950. The life of a harvestman. Nature Magazine, 43(5):264-267, 276. Blum, M. S. and A. L. Edgar. 1971. 4-methyl-3-heptanone: identification and role in opilionid exocrine secretions. Insect Biochem., 1:181-188. Cokendolpher, J. C. 1984. Homonyms of American and European Leiobunum (Opiliones, Palpatores, Leiobuninae). J. Arachnol., 12:118-19. Davis, N. W. 1934. A revision of the genus Leiobunum (Opiliones) of the United States. American Mid. Natur., 15:662-705. 1988. The Journal of Arachnology 16:126 Edgar, A. L. 1971. Studies on the biology and ecology of Michigan Phalangida (Opiliones). Misc. Publ. Mus. Zool., Univ. Michigan, No. 144, 64 pp. Edgar, A. L. and H. A. Yuan. 1968. Daily locomotory activity in Phalangium opilio and seven species of Leiobunum. Bios, 39:167-176. Ekpa, O., J. W. Wheeler, J. C. Cokendolpher and R. M. Duffield. 1985. Ketones and alcohols in the defensive secretion of Leiobunum townsendi Weed and a review of the known exocrine secretions of Palpatores (Arachnida: Opiliones). Comp. Biochem. Physiol, 81B(3):555-557. Fowler, D. J. and C. J. Goodnight. 1974. Physiological populations of the arachnid, L. longipes. Syst. Zool, 23:219-225. Holmberg, R. G., N. P. D. Angerilli and L. J. LaCasse. 1984. Overwintering aggregations of Leiobunum paessleri in caves and mines (Arachnida, Opiliones). J. Arachnol, 12:195-204. McAlister, W. H. 1962. Local movements of the harvestman Leiobunum townsendi (Arachnida: Phalangida). Texas J. Sci., 14:167-173. Mitchell, R. W. and J. R. Reddell 1971. The invertebrate fauna of Texas caves. Pp. 35-90, In Natural History of Texas Caves (E. L. Lundelius and B. H. Slaughter, eds.). Gulf Natural History, Dallas. Newman, H. H. 1917. A case of synchronic behavior in Phalangidae. Science, 45:44. Wagner, H. O. 1954. Massenansammlungen von Weberknechten in Mexiko. Z. Tierpsychol, 11:348- 352. James J. Cockerill, Department of Zoology, University of Texas, Austin, Texas 78712 USA. Manuscript received February 1987, revised June 1987. DEVELOPMENT OF PHOLCUS PHALANGI OWES (FUESSLIN) (ARANEAE, PHOLCIDAE) UNDER LONG AND SHORT PHOTOPERIODS Pholcus phalangioides (Fuesslin) commonly inhabits buildings and therefore leads a normal life even under the strong influence of human activities, especially lighting and air conditioning. In the present study the effect of photoperiod on the development of this spider was investigated to clarify the reason why it is able to settle easily in buildings. Five egg-sac-carrying females were collected from the animal rearing room of Tokyo Metropolitan University, Tokyo, in late May of 1984. The first-instar nymphs obtained from these females were kept individually in plastic vessels 5.7 cm in diameter X 11.0 cm in height for the first to third instar period and in vessels measuring 11.4 cm X 25.2 cm thereafter. A strip of thick paper was placed slantways in each vessel as a substrate. In order to investigate the effect on development of photoperiod and complete darkness, four groups each consisting of six individuals were prepared. Two groups were reared at 25° C under a long (16L-8D) or short (10L-14D) photoperiod, and two at 23.5°C under either complete darkness or 14L-10D. The light source was a 6W fluorescent tube, producing a light intensity of 250-300 lux. Relative humidity was not controlled, but fluctuated between 50 and 80%. Since first-instar nymphs molt without feeding, food supply was initiated at the second instar stage. The 2nd and 3rd- instar nymphs were alternately provided fruit flies, Drosophila melanogaster, and 1988. The Journal of Arachnology 16:127 16L-8D - Female 10L-14D - -o — o -o -o -o -o 1 6L- 8D { Male 10L-14D 50 100 150 No. of days Fig. !. — Developmental processes of Pholcus phalangioides nymphs reared at 25° C under long (16L-8D) or short (10L-14D) photoperiod. Solid circles indicate moltings, and clear ones final meltings. The horizontal line with a star indicates the male that molted six times. planthoppers, mainly Saccharosyden procerus. The number of prey provided was increased from 1-2 to 3-5 as development proceeded. The nymphs from 4th-instar to the final molt were provided with a fly, Phaenicia sericata , each time. The feeding interval was every 3 or 4 days. Experimental animals reared under complete darkness were exposed to light for a few minutes at each feeding and vessel change. Individuals dying during the rearings were omitted from the graphs subsequently obtained. Dark -Q' Female 14L100 Male 100 150 14L-10D- 50 -o No. of days Fig. 2. — Developmental processes of Pholcus phalangioides nymphs reared at 23.5° C under complete darkness as compared with that of spiders reared under 14L-10D. Symbols are the same as for Figure 1. 1988. The Journal of Arachnology 16:128 Table 1. — Growth as indicated by carapace width in Pholcus phalangioides reared at 23.5° C under 14L-10D. Figures in parentheses indicate those for adults collected from natural habitats. INSTAR ADULT 1 2 3 4 5 Female Male No. of indiv. Mean (mm) Range (mm) 15 0.63 0.59-0.66 13 0.67 0.63-0.72 8 0.98 0.91-1.03 11 1.31 1.25-1.38 5 1.75 1.66-1.88 15(12) 1.85(2.10) 1.63-2.19 (2.00-2.38) 14(12) 1.74(2.09) 1.63-1.81 (1.94-2.30) First-instar nymphs molted without feeding by the 5-6th day after emergence and showed tolerance to fasting. The mean and range of longevity for 30 fasting nymphs kept individually at 23.5° C under 14L-10D were 31.8 and 22-40 days. Figure 1 shows the result of rearings at 25° C under a long or short photoperiod. All the individuals reared under both photoperiods completed their development within a period ranging from 100 to 140 days. For females, the mean developmental period was calculated to be 116.8 days under the long photoperiod and 115.2 days under the short one. The same calculation for males gave means of 122.3 and 112.0 days, respectively. This difference of 10.3 days between the 2 photoperiods for males was insignificant (/-test). The number of molts was five for both sexes, except for one male which molted six times. In other words, development of the spiders was practically unaffected by photoperiod. According to Schaefer (1977, Z. ang. Entomol, 83:113-134), the development of many spider species is closely connected with changes in temperature and photoperiod, and reactions to photoperiod differ among species. Hamamura (1982, Japanese J. Appl. Entomol. Zool., 26:131-137) also ascertained that in Philodromus subaureolus Boesenberg et Strand, the effect of photoperiod on development was reversed before and after overwintering. Figure 2 shows the result of rearing under complete darkness as compared with that under 14L-10D, revealing that P. phalangioides developed normally even under complete darkness. In addition, the developmental period for males showed a tendency to become shorter under complete darkness than under 14L-10D. The means were calculated to be 112.5 days under complete darkness and 134.3 days under 14L-10D, but this difference of 21.8 days was statistically insignificant. According to Miyashita (1987, J. Arachnol. 15:51-58), Aehaearanea tepidariorum (C. L. Koch) developed more rapidly in darkness than under light, accompanied by a reduction in the number of molts. Table 1 shows the growth of P. phalangioides as indicated by carapace width. Measurements were performed on the individuals reared at 23.5° C under 14L- 10D. After each molt, several individuals were placed into 75% alcohol for preservation and measurement. As shown by the figures in the last column, the mean carapace width of adults was somewhat larger in individuals collected from natural habitats than in those reared artificially. The difference between them was 0.34 mm in females and 0.37 mm in males, but was not significant (/-test: P > 0.2 in the former and P > 0.05 in the latter). A similar observation was reported for Clubiona phragmitis C. L. Koch by Schaefer (op. cit.) and also for A. tepidariorum by Miyashita (op. cit.). However, the reason for this was unknown, except for the possible effect of a simple diet. 1988. The Journal of Arachnology 16:129 It can thus be concluded from the results described above that the reason why this species is able to settle easily in buildings is its apparent insensitivity to light during development. Kazuyoshi Miyashita, Department of Biology, Faculty of Science, Tokyo Metropolitan University, Fukazawa 2-1-1, Setagaya-ku, Tokyo 158, Japan. Manuscript received December 1986, revised July 1987. EGG PRODUCTION IN PHOLCUS PHALANGIOIDES (FUESSLIN) (ARANEAE, PHOLCIDAE) UNDER A CONSTANT TEMPERATURE AND PHOTOPERIOD Egg production in Pholcus phalangioides (Fuesslin) was examined. Ten final- instar female nymphs were collected from the animal rearing room of Tokyo Metropolitan University, Tokyo, in early June of 1984, and kept individually in plastic vessels 11.4 cm in diameter X 25.2 cm in height. In each vessel, a strip of thick paper was placed slantways as a substrate. After the spiders had been reared to the adult stage in the laboratory, they were mated with males obtained in the same way, and reared individually in the above-mentioned vessels until the time of death, being provided one fly, Phaenicia sericata , at intervals of 3-4 days. Five of these individuals were subjected to a second or third mating at different periods of their life, as shown in Fig. 1. An additional seven females were collected just after they had mated in natural habitats, and three of them were also subjected to a second mating in the laboratory. The time of collection was within five days after their final molt. Egg-sac production and the emergence of first-instar nymphs were recorded during the whole period of rearing, and the number of eggs per sac was determined as the number of first-instar nymphs which emerged from an egg-sac plus dead (unfertilized) eggs remaining in the sac. Dead nymphs were counted as fertilized eggs. The rearing room was maintained at 23.5° C under 14L-10D. Light was provided by 40 W fluorescent tubes, which gave a light intensity of 600-800 lux. Relative humidity was not controlled, but fluctuated within a range of between 50 and 80%. Figure 1 shows egg-sac production by 17 females. The females numbered 1-10 in the graph produced their first egg-sac 6-13 days after mating, the mean pre- oviposition period being 9.6 days. The period from oviposition to the emergence of first-instar nymphs varied from 17 to 24 days, with a mean and standard deviation (calculated from 44 egg-sacs indicated by solid circles in the graph) of 19.9+1.34 days. A large number of females continued oviposition for 400 days or more. In natural habitats, however, egg-sac-carrying females were generally found for only 100-120 days, from middle or late May to late August or early September. 1988. The Journal of Arachnology 16:130 No 1 2 3 4 5 7 8 3 10 U u _L _L 1*. _L _L_ I -o -564 -664 11 12 13 14 15 16 17 518 -O- ■593 — 677 ■522 535 1 — ' 1 * r — » 1 «■ 1 » I “ 0 100 200 300 400 500 No. of days Fig. 1. — Occurrences of egg-sac production in 17 female Pholcus phalangioides , of which 10 (nos. 1- 10) mated in the laboratory and 7 (nos. 11-17) did so in natural habitats. Solid circles indicate egg- sacs from which spiderlings emerged, and clear ones those that contained unfertilized eggs only. Arrows show matings in the laboratory. The figure following each line represents the life span of each individual in days. The number of egg-sacs produced per female varied from 2 to 8, with a mean of 4.9 sacs. Egg-sac production intervals were irregular, and showed a tendency to become longer as time passed. The total number of eggs produced per female varied from 48 to 224, with a mean of 124.5, and the mean fertilization rate was 51.5%. Therefore, the mean number of fertilized eggs produced per female was 64.1. According to the data for females numbered 6-10 and 15-17 in the graph, additional mating(s) showed almost no effect upon fertilization rate. This was considered not to have resulted from incomplete second or third matings, because the durations of the latter matings were similar to those of the first ones: mean and range of mating duration were 86.7 and 34-138 minutes in the former (n~9) and 65.9 and 22-98 minutes in the latter (71=10). The actual reason for the presence of unfertilized eggs remains unknown. As shown in Table 1, the majority of the first, second and third egg-sacs contained a mixture of fertilized and unfertilized eggs, although the percentage of egg-sacs containing only fertilized eggs was somewhat higher in the first egg-sac. The fifth and subsequent egg-sacs contained unfertilized eggs only. The total number of eggs per sac, including fertilized and unfertilized ones, gradually decreased with time. Such a tendency has been described in many species of spiders, but few authors have reported the exact state of fertility of eggs in each sac. Miyashita (1987, J. Arachnol. 15:51-58) noted that a similar tendency to that shown in Table 1 was also observed in Achaearanea tepidariorum (C. L. Koch), 1988. The Journal of Arachnology 16:131 Table 1. — Levels of egg production in relation to egg-sac sequence in Pholcus phalangioides. Figures in parentheses represent unfertilized eggs. INDIVIDUAL NUMBER EGG-SAC SEQUENCE TOTAL EGGS 1 2 3 4 5 6 7 8 Fertilized Unfertilized 1 27 30 57 0 2 29 22 32 12(10) (12) 95 22 3 28(2) (16) (10) (6) 28 34 4 20 38(2) (36) (29) (17) (5) 58 89 5 26(3) 37 47(2) 22(4) (25) (33) 132 67 6 42(1) 28(1) 26(2) 96 4 7 35(2) (35) 33(4) (27) (47) (32) (9) 68 156 8 10(2) (10) (14) (11) (16) 10 53 9 32 15(15) 5(29) (18) (28) 52 90 10 31 26 27(6) (27) (23) (28) (14) (6) 84 104 11 7(11) 7(3) 9(11) 23 25 12 34(3) 19(13) 21(8) 19(9) (31) (14) 93 78 13 27 16 19(9) (29) 62 38 14 7(11) 13(3) (18) (30) (21) 20 83 15 20(9) 18(1) 5(7) (15) 43 32 16 26(13) 25(5) 52 34(4) (35) 137 57 17 31 (47) (27) (13) (7) 31 94 X no. /sac 28.8 26.2 28.7 22.8 23.8 22.4 11.5 6.0 64.1 60.4 % fertility 88.3 66.1 60.1 27.3 0.0 0.0 0.0 0.0 although the eggs in each sac were either all fertile or all infertile, and the mean fertilization rate was rather high as compared with that in P phalangioides. However, Downes (1985, Australian J. Ecol. 10:261-264) reported that the number and fertility of eggs in Latrodectus hasselti Thorell fluctuated independently of egg-sac sequence. In the present study, the life span of adult spiders varied from 223 to 774 days, with a mean of 538.4 days. These figures are slightly lower than those for American black widow spiders, Latrodectus mactans (Fabricius), L. variolus Walckenaer and L. hesperus Chamberlin and Ivie, as reported by Kaston (1970, Trans. San Diego Soc. Natur. Hist. 16 (no. 3), 82 pp). In males reared under the same conditions as these females, adult life span varied from 132 to 277 days, with a mean of 179.8 days {n~ 11 individuals that experienced mating(s)). Kazuyoshi Miyashita, Department of Biology, Faculty of Science, Tokyo Metropolitan University, Fukazawa 2-1-1, Setagaya-ku, Tokyo 158, Japan. Manuscript received December 1986, revised July 1987. 1988. The Journal of Arachnology 16:132 INTERACTIONS BETWEEN THE CRAB SPIDER M1SUMENA VATIA (CLERCK) (ARANEAE) AND ITS ICHNEUMONID EGG PREDATOR TRYCHOSIS CYPERIA TOWNES (HYMENOPTERA) Spider egg masses are subject to a wide variety of dangers, including insects whose larvae require them as a food source. Information on these spider-insect relationships often consists largely of documenting predator and prey species (Askew 1971), and in many instances even predator or parasitoid records are missing (Krombein et al. 1979; Austin 1985). In my study area along the coast of Maine, U.S.A., the ichneumonid wasp Trychosis cyperia Townes (Hymenoptera: Ichneumonidae) is an egg predator on the crab spider Misumena vatia (Clerck) (Araneae: Thomisidae). I define an egg predator as an individual that attacks the eggs of a mass collectively and feeds externally on it, rather than developing within a single egg oviposited there by its parent (Austin 1985). A single larva of T cyperia will totally consume all but the largest of Misumena egg masses, before pupating within the spider’s nest (Morse and Fritz 1987). Trychosis is a potentially important egg predator on Misumena , since one successful attack usually totally destroys the spider’s entire reproductive effort (Morse and Fritz 1987). Further, Trychosis may successfully attack between 7 and 60% of the Misumena nests in a local population (Morse and Fritz 1987; Morse, unpublished data). As a result of Trychosis ’ high predation level, Misumena should experience strong selection to minimize wasp attacks. Indeed, female Misumena guard their egg masses over much or all of the period between egg-laying and emergence of the young from the nest about a month later (Morse 1985). Predation by these wasps is not random: they successfully attack small egg masses, which are guarded by small spiders, significantly more frequently than large ones. This pattern is a consequence of differences in guarding behavior by different-sized spiders. Nests from which the parents are removed do not differ in success as a consequence of egg mass size (Morse, unpublished data). This result strongly suggests that the differential predation is a consequence of direct interactions between Misumena and Trychosis , in which large spiders fare better than small. However, although I monitored the nesting success of over 200 spiders at three different sites between 1982 and 1985, I did not observe Trychosis adults in the field, even though predation by it was sometimes high. During the summer of 1986, I finally observed Trychosis at Misumena nests, and the response of guarding Misumena to them. I have been unsuccessful in finding reports of similar interactions in the literature, and therefore describe them in detail, both to document their characteristics and to draw them to the attention of others who might be in a position to observe similar behavior. In the first observation, a brief encounter, a Trychosis landed on the upper surface of a Misumena nest, located in a turned-under leaf, 40 cm up a milkweed plant (see Morse 1985 for a description of Misumena nests). After a few seconds, it moved out of view over the side of the nest to the under surface, flicking its wings and abdomen rapidly. It encountered the guarding female crab spider and 1988. The Journal of Arachnology 16:133 instantly flew from the nest and out of sight. The spider’s front legs were raised at the instant after the wasp left, a pattern I have otherwise only observed when a spider is ready to strike at prey (Morse 1979). The spider lowered her legs within 10 sec. This ichneumon did not probe the nest with its ovipositor while it was within sight; indeed the brood it visited was 19 days old; therefore, the spiderlings inside probably were nearly ready to molt into their second instar, and it seems unlikely that an egg predator would be able to exploit this nest successfully. This nest was not parasitized, and young eventually emerged from it. This female weighed 74 mg after egg-laying, near the average mass for post-reproductive females of this population in 1986 (X ± SD = 76.7 ± 20.4 mg., N = 171). The second encounter was much more protracted, and involved an eight-day- old next guarded by an extremely large female spider (114 mg). I initially observed the ichneumon on a leaf 45 cm above the ground, near the top of a milkweed plant, three cm from an adjacent leaf with a Misumena test. Initially the wasp was largely stationary, although its antennae remained in constant motion. At that instant the spider occupied the underside of her nest out of the direct line of vision from the wasp. After 30 sec the ichneumon became active and walked about in a tight circle for about 30 sec before taking its previous position. Two mintues later it moved to the underside of its leaf. During this period the spider was extremely active for a guarding individual (see Morse 1987). It moved to the top of its nest and subsequently changed position 14 times over the next 30 min. These movements included both shifts between the underside and upperside, and between the petiole of the nest leaf and the nest at the terminal end of this leaf. This rate is eight times greater than that of average guarding spiders at other times. ( X ± SD - 3.3 ± 3.8 moves/ h, virtually all associated with nest maintenance; N - 34: Morse, unpublished data), and twice the rate of the most active guarding Misumena I have monitored. Approximately 30 minutes from the beginning of these observations, the ichneumon walked to the upper surface of the nest from the leaf it had previously used. At this time the spider occupied the upper surface of its nest, on the distal end of the leaf. As the wasp neared the nest from the proximal end, the spider instantly became active. It approached and attacked the wasp, seemingly as it would attack a prey item, raising its front pairs of legs and striking down on it. However, the spider did not bite the wasp; instead, it flung the wasp from the nest toward the ground. The wasp landed on my trousers leg, a few cm distal to the nest leaf and about 30 cm below the nest. It remained there for one minute, behaving as it did on the originally-occupied leaf, largely stationary, but regularly moving its antennae. Perhaps this initial action would normally have sufficed to remove the wasp from the vicinity of the nest. I then picked the wasp up on a blade of grass and placed it back on the spider’s upper nest surface. The spider again quickly attacked, but this time it only displaced the wasp 2-3 cm; the wasp landed on the extreme distal end of the nest following this attack. Instantly the wasp moved to the side of the nest and walked rapidly along the side, probing several times with its ovipositor. Post-reproductive spiders draw the upper and lower parts of their nests together tightly with silk; but this junction might nevertheless provide the most satisfactory place to insert an ovipositor. An ovipositor thrust into the top or bottom of the nest would penetrate the milkweed leaf, and thus run the risk of becoming clogged by milkweed latex (see Dussourd and Eisner 1987). 1988. The Journal of Arachnology 16:134 The spider attacked the wasp in a similar way on the side of the nest, but ineffectually, for the wasp merely retreated to the other side of the nest and probed there with its ovipositor. The spider moved toward the wasp once again, but the wasp ran directly over the spider and across the dorsal side of the nest, with its ovipositor pointed downward. The spider was on the opposite side of the nest at this time. The wasp again moved to the side of the nest and inserted, or attempted to insert, its ovipositor between the upper and lower layers of the nest. After this action, it moved to the leaf it had occupied at the beginning of the observations. During this entire period it did not fly. After it had remained largely inactive on its original site for 10 minutes, I again placed it on the spider’s nest, and the spider once more attacked it. The wasp retreated this time, and moved to another adjacent leaf. I then attempted to place the wasp on another spider’s nest, but upon being placed there, it flew for the first time, and I soon lost sight of it. Given the aggressive response of the spiders, it may seem surprising that the second spider did not kill the wasp. This initial aggressive response was similar to the one I have observed when these spiders attack other hymenopterans. Since I have seen even post-reproductive Misumena with yellowjacket ( Vespula sp.) kills (Morse 1987), they must be capable of penetrating Try cho sis’ carapace. Further, both pre- and post-reproductive Misumena regularly take small euminid wasps, insects of a comparable size and carapace hardness to Trychosis. Although reproductive spiders do not actively seek food, some of them do capture occasional insects that approach them while they guard their nests (Morse 1987). The initial response of the spiders may typically suffice to dissuade these egg predators, as the first wasp’s behavior suggests. Further, physically displacing Trychosis from the nest may normally keep it from attacking again. Askew (1971) notes that caterpillars may regularly elude ichneumonid parasitoids by descending from a leaf on a thread, and that ichneumonid pupae may themselves escape pteromalid hyperparasites by dropping into the substrate. These observations suggest that some hymenopterans are not highly skilled at tracking mobile targets; perhaps they also experience difficulty in relocating a stationary target from which they have been displaced. The interactions at the second nest suggest a possible explanation for the difference in predation levels on the egg masses of large and small spiders. The large spiders may on average be more successful in quickly removing the wasps from their nests than are the small spiders. Subsequent efforts of the spider at the second nest became progressively less effectual. Nevertheless the ichneumonid did not successfully attack the second nest, because no wasp offspring emerged. I could not determine whether the wasp actually laid an egg, however. It may require a minimum period on the nest to determine whether the nest is a satisfactory egg-laying site. The fact that the first wasp visited a nest probably far too old for it to exploit suggests that Trychosis ’ initial level of discrimination is low. I thank J. K. Waage for comments on the manuscript and C. S. Hieber for helpful discussion. H. Townes identified Trychosis. My research on crab spider reproductive ecology has been supported by the National Science Foundation (BSR85-16279). 1988. The Journal of Arachnology 16:135 LITERATURE CITED Askew, R. R. 1971. Parasitic Insects. Heinemane Educational Books, London. 316 pp. Austin, A. D. 1985. The function of spider egg sacs in relation to parasitoids and predators, with special reference to the Australian fauna. J. Natur. Hist., 19:359-376. Dussourd, D. E. and T Eisner. 1987. Vein-cutting behavior: insect counterplay to the latex defense of plants. Science, 237:898-901. Krombeie, K. V., P D. Hurd, Jr., D. R. Smith and D. B. Burks, editors. 1979. Catalog of Hymenoptera in America North of Mexico. Smithsonian Institution Press, Washington, D. C. Morse, D. H. 1979. Prey capture by the crab spider Misumena calycina (Araneae: Thomisidae). Oecologia, 39:309-319. Morse, D. H. 1985. Nests and nest-site selection of the crab spider Misumena vatia (Araneae: Thomisidae). J. Arachnol., 13:383-390. Morse, D. H. 1987. Attendance patterns, prey capture, changes in mass, and survival of crab spiders Misumena vatia (Araneae, Thomisidae) guarding their nests. J. Arachnol, 15:199-210. Morse, D. H. and R. S. Fritz. 1987. The consequences of foraging for reproductive success. Pp 443- 455, In Foraging Behavior (A. C. Kamil et aL, eds.). Plenum, New York. Douglass H. Morse, Graduate Program in Ecology and Evolutionary Biology, Division of Biology and Medicine, Brown University, Providence, Rhode Island 02912 USA. Manuscript received July 1987, revised October 1987. 1988. The Journal of Arachnology 16:136 BOOK REVIEWS Roberts, M. J. 1987. The Spiders of Great Britain and Ireland. Harley Books, Martins, Great Horkesley, Colchester, Essex, C06 4AH England. Vol. 2 (Linyphiidae) (Price £45.00). The first and third volume of this three volume work were reviewed by B. J. Kaston in this Journal (Vol. 13, no. 2, pp. 275-276) in 1985. Volume two has now been published, and includes descriptions and illustrations of 267 species of linyphiids, placed in 105 genera. There is also an addendum listing six nonlinyphiid spiders new to Great Britain, a page and half illustrating variation in Araneus diadematus , a glossary, a checklist of British spiders, and an index to scientific names. The short key segregates the linyphiid species into four groups on the basis of tibial spines and metatarsal trichobothria. The key is followed by four tables, one for each group, listing genera and species and comparing the general appearance, total length of the specimens, positions of the trichobothria and tibial spines, with reference to the text figures. It is hoped that the key and tables, with the help of the illustrations, will guide the reader to a correct determination. The volume is lavishly illustrated. For most species, one page has the species name, references to illustrations, plus a few lines of description, and the illustrations are on the facing page. A paragraph under the last species of each genus summarizes the distinguishing characters of the species included within the genus, while another paragraph summarizes distribution and habitats. There are no literature citations other than the original description of the species. The epigynes are illustrated next to the palpi, with views of carapaces on a separate page. Unlike the previous volumes, Volume 2 has all illustrations to the same scale, genitalia 90X, carapaces 60X. Thus many illustrations are of magnificent size, while illustrations of smaller species may be barely visible. Many illustration pages show an unusual amount of blank space due to the minute size of the species and illustrations. There are several ways to illustrate spider genitalia. In North America illustrations are prepared using reflected light at the time the genitalia are examined. Much less satisfactory are illustrations of genitalia mounted on a microscope slide and examined by transmitted light. Slide mounted palpi are difficult to place in comparable positions, they may be compressed, they deteriorate rapidly (when not dismounted), and often such slides become separated from the specimen. Unfortunately, however, this method is popular because compound microscopes are more readily available than stereoscopic dissecting microscopes. The third and most modern method for illustrating spider genitalia is by use of the scanning electron microscope (SEM). While SEMs show minute details of surface sculpturing, and may be extraordinarily interesting and valuable, it is difficult to use this method for comparison with specimens to be keyed out. Roberts has illustrated with reflected light, the best method for the 1988. The Journal of Arachnology 16:137 purpose. Often two or more epigyna are illustrated for the same species, showing individual differences due to variable transparency of the surface. It might have been more useful if Roberts had provided only one illustration of an unprepared epigynum and, for the second, had used a cleared epigynum, showing the underlying ducts. This new volume should be compared with its counterpart, vol. 2 of Locket, G. H. and A. F. Millidge’s British Spiders, Ray Soc. 1953. Millidge illustrated only 250 species, included a key to linyphiid genera (but not to species) and gave citations for species in addition to the original ones. Roberts’ illustrations usually show much more detail, especially of the epigyna. However, Robert’s illustrations of small species are smaller than those by Millidge. If there was criticism of Millidge’s work, it was that illustrations of epigyna, palpi and carapaces of the same species were often on separate pages and that the palpi were all illustrated in retrolateral (lateral) view. While the lateral view may give most diagnostic characters for the limited fauna of the British Isles, it makes their interpretation impossible for an outsider who wants to study the palpus of a given genus; for this purpose, a ventral view would be needed in addition to the lateral. This limitation makes Millidge’s illustrations frustrating to use for North American or Holarctic species that need to be placed into genera. Unfortunately Roberts illustrates the same lateral view of the palpus that Millidge used, albeit with greater detail and, in the larger illustrations, finer craftsmanship (however, some small illustrations show less detail). Millidge’s treatment frequently gives a second illustration of the palpal tibia to assist the identification, but there is only one palpus illustration in Roberts’ volume. In summary: this is a superb, authoritative volume. Perhaps inevitably, it will be more useful to those working with the British fauna than to those who want to increase their understanding of linyphiid spiders of the world. Herbert W. Levi, Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138 USA. Eberhard, W. G., Y. D. Lubin and B. C. Robinson (Eds.). 1986. Proceedings of the Ninth International Congress of Arachnology, Panama 1983. Smithsonian Institution Press, Washington, DC. 334 pp. (Price $25). One of the most difficult tasks a reviewer can be assigned is the evaluation of a proceedings volume from a general meeting. There is no unifying theme in such a book except for the limitations on the membership of the sponsoring group (here all the papers concern arachnids, though insects are the real focus of at least two of them). The erratic quality of the short papers, many serving as abstracts of a more complete article published elsewhere, is particularly obvious when there has been no pre-presentation screening. It will come as no surprise to those who have read several such volumes that a small number of the papers probably could not have been published in a reviewed journal. The less said about these efforts, the better. The largest class of papers consists of reports of the smaller byways and peculiar backwaters explored during the author’s main research efforts, or first attempts by students, or, sad to say, 1988. The Journal of Arachnology 16:138 the same paper presented at the previous meeting, or the one before that, with a few cosmetic alterations. A majority of the papers in this volume, however, are interesting and valuable nonetheless. Bleckmann (p. 19) presents an elegant analysis of the response of Dolomedes spiders to surface waves on water, showing how the spiders discriminate waves caused by prey and extract a surprising amount of information from them. For me, this was the outstanding paper in the volume. New behavioral phenomena and new structures connected with them are reported in the papers by Coyle (p. 33) on mating in Euagrus (the males use a patch of spines that functions like Velcro™) and by Robinson, Robinson, Murphy, and Corley on egg-sac burying by Nephila maculata. These papers are refreshingly original and well written and illustrated. Edmunds (p. 61) uses a detailed study on the stabilimentum in two species of Argiope to review stabilimentum function and evolution in orb weavers in general, concluding, sensibly, that stabilimentum function may vary from species to species. In the area of systematics and biogeography, van Helsdingen’s (p. 121) survey of the world distribution of Linyphiidae provides an important data base and should set the course of systematic research in this neglected family for some time to come. Quintero (p. 203) presents a new classification of Amblypygi which may prove controversial but which is well argued and amply illustrated. Finally, Raven (p. 223) summarizes his new treatment of the mygalomorphs, the details of which have now been published elsewhere. The editing of this volume, unfortunately, leaves much to be desired, but primarily on the technical side — there should have been more careful proofreading. It is particularly bothersome to have obvious typographical errors in the boldface titles of articles. For example, on p. 301, the word “Summary” is treated as if it were the name of the author of a species. On p. 320 we see “Hersilidae” instead of Hersiliidae. On p. 332, “Argiope bruennichii” is given as the name of a coauthor of a paper. Even in the table of contents one finds words like “umarobiid.” Throughout there is erratic application of the convention of putting species names in italics, including one of the papers by the senior editor. A few of the illustrations (see pp. 183 and 186) are not of publishable quality but the blame here must be shared with the authors. An organization of the papers and abstracts into biological categories would have been preferable to publishing them in alphabetical order by the author’s last name. However, the price of the volume is reasonable and the majority of the papers are worth reading. Despite the several sour chords struck in the paragraphs above, I recommend that professionals in the field add it to their personal libraries. I also recommend that in future such volumes not be published. The required limitations on the included papers, their erratic quality, the tanatalizing nature of abstracts that stand alone (and in several cases, as of this writing, reports abstracted in this book have not yet appeared, some four years later), are significant shortcomings. Add to this the difficulties of finding such “one-shot” volumes in libraries. Of the three congresses or meetings I attended in 1987, the most valuable (a Smithsonian-sponsored conference on the evolution of terrestral ecosystems) banned the presentation of papers and instead organized the participants into overlapping working groups charged with summarizing the past, present, and 1988. The Journal of Arachnology 16:139 future of an area of research in an informal report. How much more refreshing and stimulating it would be if the participants in congresses and meetings simply discussed their current research in a less formal, more open, and speculative fashion, without the constraints of having to present a finished paper for publication. I vote for talks about research in progress, followed by vigorous discussion, rather than formal papers on last year’s results! William A, Shear, Department of Biology, Hampden-Sydney College, Hampden-Sydney, Virginia 23943 USA. THE AMERICAN ARACHNOLOGICAL SOCIETY President: William A. Shear (1987-1989) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 Membership Secretary: Norman I. Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: James W. Berry (1987-1988) Department of Zoology Butler University Indianapolis, Indiana 46208 Directors: James C. Cokendolpher (1987-1989), W. G. Eberhard (1986-1988), Jerome S. Rovner (1987-1989). Honorary Members: P. Bonnet, W. J. Gertsch, H. Homann, R. F, Lawrencef, H. W. Levi, G. H. Locket, A. F. Millidge, M. Vachon, T. Yaginuma. The American Arachnological Society was founded in August, 1972, to promote the study of Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachnology. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $25.00 for regular members, $15.00 for student members. Correspondence concerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addition, members receive the bi-annual newsletter of the Society, American Arachnology. American Arachnology, edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-find literature, information about arachnology courses and professional meetings, abstracts of papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arachnology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. President-Elect: George W. Uetz (1987-1989) Department of Biological Sciences University of Cincinnati Cincinnati, Ohio 45221 Treasurer: Gail E. Stratton (1987-1989) Department of Biology Albion College Albion, Michigan 49224 Archivist: Vincent D. Roth Box 136 Portal, Arizona 85632 Research Notes Dragonfly predation upon Phidippus audax (Araneae, Salticidae), Orrey P. Young and Timothy C. Lockley. 121 Notes on aggregations of Leiobunum (Opiliones) in the southern U. S. A., James J. Cockerill 123 Development of Pholcus phalangioides (Fuesslin) (Araneae, Pholcidae) under long and short photoperiods, Kazuyoshi Miyashita 126 Egg production in Pholcus phalangioides (Fuesslin) (Araneae, Pholcidae) under a constant temperature and photoperiod, Kazuyoshi Miyashita 129 Interactions between the crab spider Misumena vatia (Clerck) (Araneae) and its ichneumonid egg predator Trychosis cyperia Townes (Hymenoptera), Douglass H. Morse 132 Book Reviews The Spiders of Great Britain, by M. J. Roberts, Herbert W. Levi 136 Proceedings of the Ninth International Congress of Arachnology, Panama 1983, edited by W. G. Eberhard, Y. D. Lubin and B. C. Robinson, William A. Shear 137 "'CONTENTS fee.;.. \ ^ 8? THE JOURNAL OF ARACHNOLOGY v> . • ^ • y« •- «* .J& • ~ . VOLUME ft*. J'“S. . Feature Articles NUMBER 1 . *'iSvA*..i. .‘yQ ^x .xy • V '*”>• r.' «'%»•> " *. •::,•■ » ' N> * »> Silk use during mating in. Pisaurina mira (Walckenaer) (Araneae, Pisauridae), John A. Bruce and James E. Carico 1 Ecologia y aspectos del comportamiento en Linothele sp. (Araneae, Dipluridae), Nicolas Paz S 5 Contribution al conocimiento taxonomico del genero Urophonius Pocock, 1893 (Scorpiones, Bothriuridae), Luis Eduardo Acosta. 23 Interspecific tolerance in social Stegodyphus spiders (Eresidae, Araneae), U. Seibt and W. Wickler 35 The effect of temperature on oviposition interval and early development in Theridion rufipes Lucas (Araneae, Theridiidae), Michael F. Downes. 41 The spider genus Paratheuma Bryant (Araneae, Desidae), Joseph A. Beatty and James W. Berry 47 Spiders (Araneae) associated with strip-clearcut and dense spruce-fir forests of Maine, Daniel T. Jennings , Mark W Houseweart , Charles D. Dondale and James H. Redner 55 Triaenonychidae Sudamericanos. III. Description de los nuevos generos Nahuelonyx y Valdivionyx (Opiliones, Laniatores), Emilio A. Maury 71 Spiders (Araneae) captured in Malaise traps in spruce-fir forests of west-central Maine, Daniel T. Jennings and Daniel J. Hilburn 85 Reproductive periods of Phidippus species (Araneae, Salticidae) in South Carolina, Steven H. Roach 95 The spider genus Cybaeota (Araneae, Agelenidae), Robert G. Bennett 103 (continued on back inside cover) Cover photograph, spider figure on men’s meeting house, Palau, by J. W. Berry Printed by the Texas Tech Press, Lubbock, Texas, U.S.A. Posted at Lubbock, Texas June 10, 1988 The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNQLOGICAL SOCIETY VOLUME 16 SUMMER 1988 NUMBER 2 THE JOURNAL OF ARACHNOLOGY EDITOR: James E. Carico, Lynchburg College ASSOCIATE EDITOR: Jerome S. Rovner, Ohio University EDITORIAL BOARD: J. A. Coddington, National Museum of Natural History, Smithsonian Institution; J. C. Cokendolpher, Texas Tech University; F. A. Coyle, Western Carolina University; C. D. Dondale, Agriculture Canada; W. G. Eberhard, Universidad de Costa Rica; M. E. Galiano, Museo Argentino de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia, Missouri; N. F. Hadley, Arizona State University; N. V. Horner, Midwestern State University; H. W. Levi, Harvard University; E. A. Maury, Museo Argentino de Ciencias Naturales; M. H. Muma, Western New Mexico University; N. I. Platnick, American Museum of Natural History; G. A. Polis, Vanderbilt University; S. E. Riechert, University of Tennessee; A. L. Rypstra, Miami University, Ohio; M. H. Robinson, U.S. National Zoological Park; W. A. Shear, Hampden-Sydney College; G. E. Stratton, Albion College; W. J. Tietjen, Lindenwood College; G. W. Uetz, University of Cincinnati; C. E. Valerio, Universidad de Costa Rica. THE JOURNAL OF ARACHNOLOGY (ISSN 0161-8202) is published in Spring, Summer, and Fall by The American Arachnological Society at Texas Tech Press. Individual subscriptions, which include membership in the Society, are $30.00 for regular members, $20.00 for student members. Institutional subscriptions to The Journal are $50.00. Correspondence concerning subscriptions and member- ships should be addressed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. Susan E. Riechert, Department of Zoology, University of Tennessee, Knoxville, TN 37916 U.S. A., at $15.00 for each number, $40.00 for a complete volume; the Index to Volumes 1-10 is available for $10.00. Remittances should be made payable to The American Arachnological Society. Correspondence concerning undelivered issues should be addressed to the Texas Tech Press, Texas Tech University, Lubbock, Texas 79409 U.S. A. Change of address notices must be sent to the Membership Secretary. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without benefit of review. Manuscripts and all related correspondence must be sent to Dr. Jerome S. Rovner, Associate Editor, Department of Zoological Sciences, Ohio University, Irvine Hall, Athens, Ohio 45701 U.S. A. PROOFS, REPRINTS, and CHARGES: Authors will receive a reprint order form along with their proofs. Reprints are billed at the printer’s current schedule of costs. All authors will be billed by the printer for page charges. The charge per journal page will be assessed as follows: $30. 00-nonmembers; $25. 00-members acknowledging grant, institutional, or other support; $ 15.00-full members who do not have support; $ 10.00-student members who do not coauthor with full members and who do not acknowledge support. Members may have page charges reduced for a limited number of pages; a statement of justification for reduction should accompany the manuscript when it is first sent to the Associate Editor for review. Viera, V. y F. G. Costa. 1988. Analisis del comportamiento de captura de presas por machos adultos de Metepeira sp. A (Araneae, Araneidae), utilizando telas de juveniles y hembras adultas coespecificos. J. Arachnol., 16:141-152. ANALISIS DEL COMPORTAMIENTO DE CAPTURA DE PRESAS POR MACHOS ADULTOS DE METEPEIRA SP. A (ARANEAE, ARANEIDAE), UTILIZANDO TELAS DE JUVENILES Y HEMBRAS ADULTAS COESPECIFICOS U'BR ARIES Carmen Viera y Fernando G. Costa Division Zoologia Experimental Instituto de Investigaciones Biologicas Clemente Estable Av. Italia 3318, Montevideo, Uruguay ABSTRACT Prey ( Acromyrmex sp. ants) were offered to four experimental groups of Metepeira sp. A: adult females on their own webs; juveniles on their own webs; adult males on webs built by adult females; adult males on webs built by juveniles. Although adult males do not build webs, they are fully capable of capturing prey in foreign webs. Male behavior was basically similar to females and juveniles. However, males showed high frequency in certain units of behavior in the prey detection phase (particularly males on female webs). A sexual origin for these units of behavior is suggested. Sexual interference would reduce the predatory efficiency of males but also probably reduce the risk of female predation on males in the field. RESUMEN Se entregaron presas (hormigas Acromyrmex sp.) a cuatro grupos experimentales de Metepeira sp. A: hembras adultas ocupando sus telas; juveniles ocupando sus telas; machos adultos ocupando telas de hembras; machos adultos ocupando telas de juveniles. Los machos adultos, que no construyen telas orbiculares, mostraron plena habilidad predatoria en telas ajenas. El comportamiento de los machos fue basicamente similar a hembras y juveniles. Sin embargo, los machos presentaron alta frecuencia de algunas unidades de comportamiento en la fase deteccion (principalmente machos en telas de hembras). Se sugiere un origen sexual de estas unidades. Esta caracteristica afectaria la eficiencia de captura pero disminuiria los riesgos de predation sobre los machos en el campo. INTRODUCCION Los machos adultos de Araneidae pierden la capacidad de construir telas orbiculares a partir de su muda de maduracion y consecuentemente no se alimentarian (Bristowe 1941; Millot 1949; Foelix 1982). Sin embargo, en especies de Nephila y Argiope , los machos son mucho mas pequenos que las hembras, ocupan y eventualmente se alimentan en telas de hembras adultas o en penultimo estadio (Christenson y Goist 1979; Robinson y Robinson 1978; Vollrath 1980; Christenson 1984; Christenson et al 1985). En Eriophora fuliginea (C. L. Koch) los machos tienen un tamano similar a las hembras, construyen redes orbiculares y capturan presas (Robinson et al. 1971; Robinson y Robinson 1981). Eberhard et al. (1978) senalaron robos de telas de Metazygia gregalis (O.R-Cambridge) por 142 THE JOURNAL OF ARACHNOLOGY machos adultos de varias especies orbitelares, incluyendo machos coespecificos, utilizandolas para capturar presas. Pese a estos antecedentes, no se conocen descripciones detalladas del comportamiento de captura de presas por machos adultos de Araneidae. Las telas de las especies de Metepeira se caracterizan por la presencia de un refugio y de hilos de conexion con el centro de la tela orbicular. Los machos adultos tienen un tamano semejante a las hembras y pueden utilizar telas ajenas para capturar presas. Por ejemplo, los machos de M. grinnelli (Coolidge) pueden desplazar a individuos de Cyclosa turbinata (Walckenaer), ocupar sus telas y capturar presas (Spiller 1984). Tambien los machos adultos de Metepeira sp. A ocupan y capturan presas en telas de juveniles y hembras coespecificos (Viera y Costa 1985). Este ultimo resultado nos estimulo a averiguar si el comportamiento de captura de presas de los machos adultos de esta especie es similar al descripto por Viera (1986) para los juveniles y las hembras adultas. Los objetivos de este trabajo son: (i) Describir el comportamiento de captura de presas por machos adultos de Metepeira sp. A ubicados en telas de juveniles y hembras coespecificos; (ii) Describir el comportamiento de captura de presas por juveniles y hembras adultas de Metepeira sp. A (grupos de control); (iii) Analizar comparativamente los comportamientos de captura de juveniles, hembras, machos en telas de juveniles y machos en telas de hembras, sobre una misma presa. Este estudio constituye el primer intento de analizar el modelo comportamental de captura de presas, realizado por machos adultos de Araneidae. Los resultados tambien permitiran evaluar la capacidad depredadora de los machos y la influencia del tipo de tela utilizada, asi como vincular esta actividad alimentaria con las tacticas reproductoras de Araneidae. MATERIAL Y METODO Se colectaron 145 individuos juveniles y adultos de Metepeira sp. A (denomination provisoria, sugerida por H. W. Levi, Harvard University) en Punta Espinillo, Montevideo, Uruguay. Las telas fueron localizadas en infloraciones y parte superior del tallo de Eryngium sp. (Umbelliferae), con el refugio ubicado generalmente entre la umbela y el pedicelo (hasta cinco telas en el mismo tallo). Todos los individuos utilizados fueron depositados en la coleccion aracnologica del Museo Nacional de Historia Natural, Montevideo (lote N° 305). En el laboratorio los individuos fueron criados en frascos individuates de 9 cm de diametro y 14 cm de altura, con un recipiente con agua y un soporte para la tela, cerrados con una malla de nailon. Fueron alimentados con larvas de Tenebrio sp. (Coleoptera). La temperatura media durante 166 dias de cria y estudio fue 22.96 ± 1.95°C. Para las experiencias los individuos se trasladaron a cajas de vidrio de 30 X 30 X 9 cm, con un marco interno de madera y un recipiente con agua. Las arahas que mudaron se usaron despues de cinco dias. Las presas elegidas fueron obreras de Acromyrmex sp. (Hymenoptera, Formicidae) con fuertes defensas mecanicas y muy abundantes en el sitio de colecta. El tamano de la presa fue igual o ligeramente inferior a la arana, ubicandola en la zona inferior de la tela, 2 h despues de su captura. Para la observation se utilize un fondo oscuro, una luz puntiforme lateral de 445 lux y una lupa amplia de 2X. Se relataron y registraron las observaciones con un grabador magnetofonico, se filmaron algunas secuencias con una camara VIERA Y COSTA— CAPTURA DE PRESAS POR MACHOS DE METEPEIRA 143 cinematografica Super-8 y se analizaron cuadro a cuadro en una moviola. Las secuencias comportamentales se describieron separando unidades y fases comportamentales de acuerdo a Robinson y Olazarri (1971) y particularmente Viera (1986). El diseno experimental fue el siguiente: Experiencia 1 : Se colocaron hembras adultas individualmente en los recipientes de experimentacion iimpios. A las 24 6 48 h se entrego la presa a las aranas que construyeron tela y se observe y registro la captura. Experiencia 2: Se colocaron machos adultos en recipientes de experimentacion inmediatamente despues de ser extraidas las hembras adultas (sin alterar la tela). Veinticuatro horas despues se controlo la ocupacion de la tela por el macho, se entrego la presa y se observo y registro la captura. Experiencia 3: Se siguio el mismo procedimiento de la Experiencia 1, pero usando juveniles (en penultimo y antepenultimo estadios) en sustitucion de las hembras. Experiencia 4: Se siguio el mismo procedimiento de la Experiencia 2, pero colocando machos en telas construidas por individuos juveniles. Las observaciones correspondientes a las cuatro experiencias se realizaron intercaladas entre si. Se registraron las experiencias desde el momento de colocar la presa ( inicio ). Las sucesiones de unidades se contabilizaron a partir de la primera unidad de comportamiento que se observo despues de entregar la presa, exceptuando el reposo anterior de la arana. Se considero como fin de las experiencias la ingestion o abandono de la presa, asi como tambien el mantenimiento de quietud o acicalamiento por un lapso mayor a 60 seg. La eficiencia de captura se calculo: (N° de presas capturadas/N° de presas etregadas) x 100. La temperatura media durante las observaciones experimentales fue: 23.3 ± 2.3° C (N = 78). Los resultados obtenidos de las cuatro experiencias se compararon cualitativamente y cuantitativamente. No se compararon entre si las experiencias 1 y 4, ni las experiencias 2 y 3, para evitar la incidencia simultanea de dos variables (estado fisiologico y medio ambiente experimental). Se utilizaron los estadisticos: test de probabilidad exacta de Fisher, test de dos muestras de Wilcoxon (Siegel 1956) y test de diferencia de medias de Student con restricciones para la varianza (paquete PRESTA, Centro S. Ramon y Cajal, Espana). El nivel minimo para rechazar la hipotesis nula fue 0.05. RESULTADOS El comportamiento de captura en los cuatro grupos experimentales mostro un patron comun, donde las unidades de comportamiento se sucedieron en el tiernpo. De acuerdo con Viera (1986), se reconocieron ties fases: (1) Fase de deteccibn de la presa , constituida por las unidades desplazamiento (locomotion de la arana en la tela), tensamiento (tironeo de los radios de la tela) y toqueteo (golpes suaves de patas sobre la presa); (2) Fase de inmovilizacion de la presa , consituida por las unidades envolvimiento (sujecion con ataduras de seda), mordeduras cortas (inserciones sucesivas de los queliceros en la presa) y mordedura prolongada (insertion de los queliceros en la presa durante 20 segundos como minimo); (3) Fase terminal , constituida por las unidades transporte (liberation y traslado de la presa en las patas IV hasta el lugar de ingestion) y manipulation de la presa (maniobras de ubicacion de la presa, previas a la ingestion). 144 THE JOURNAL OF ARACHNOLOGY Se observaron otras dos unidades: quietud (inmovilidad total) y acicalamiento (limpieza aparente de los apendices). Quietud se vinculo frecuentemente con la fase inmovilizacion y acicalamiento se vinculo con deteccion e inmovilizacion. Sucesion de unidades de comportamiento de hembras (Experieneia 1). — De 21 hembras empleadas, 19 hicieron tela. Diecisiete iniciaron la captura desplazandose hasta el centre de la tela (Fig. 1), mientras que dos hembras partieron desde el centro de la tela (una de estas no construyo refugio). Desde el centro todos los individuos se desplazaron directamente hacia la presa, salvo uno que previamente realizo tensamiento. Un solo individuo efectuo toqueteo sobre la presa, despues del desplazamiento inicial. Dieciocho individuos envolvieron la presa inmediatamente despues de desplazarse. La fase inmovilizacion mostro una fuerte vinculacion entre envolvimiento y mordeduras cortas. Ambas unidades se vincularon en menor medida con mordedura prolongada (Fig. 1). La fase terminal se inicio con la unidad transporte, sucediendo a envolvi- miento, mordedura prolongada o excepcionalmente mordeduras cortas (2 individuos). Diecisiete hembras presentaron fase terminal e ingirieron en el refugio; dos hembras no realizaron fase terminal y finalizaron el comportamiento en quietud. Quietud se relaciono fundamentalmente con envolvimiento; tambien se vinculo con mordedura prolongada, mordeduras cortas, acicalamiento y transporte. Acicalamiento se observo repetidamente en un solo individuo, relacionado con envolvimiento y quietud. La eficiencia de captura de presas fue 95% (un solo individuo no capture la presa). Sucesion de unidades de comportamiento de machos que utilizaron telas de hembras (Experieneia 2). — De 18 machos empleados, 17 ocuparon telas de hembras. Quince individuos iniciaron el comportamiento de captura desde el refugio y dos desde el centro de la tela (una tela no poseia refugio). Once machos realizaron como primera unidad desplazamiento y seis tensamiento (Fig. 2). Ocho machos realizaron tensamiento y 10 machos realizaron toqueteo durante la fase deteccion de la presa. La fase inmovilizacion se inicio con las sucesiones desplazamiento a envolvimiento o toqueteo a envolvimiento. Envolvimiento se relaciono estrecha- mente con mordeduras cortas y ambas unidades se vincularon con mordedura prolongada (Fig. 2). Ocho machos comenzaron la fase terminal con la sucesion envolvimiento a transporte y un individuo con la sucesion mordedura prolongada a transporte. Quietud se vinculo con mordeduras cortas y mordedura prolongada. Cinco individuos terminaron el comportamiento realizando quietud (dos de ellos ubicados en el refugio) y uno desplazandose fuera de la tela. Acicalamiento fue observado en cinco individuos, relacionandose principalmente con desplaza- miento y en menor medida con quietud, envolvimiento y tensamiento. Un individuo termino su comportamiento en acicalamiento prolongado. La eficiencia de captura de presas fue 72% (cinco machos no capturaron las presas). Sucesion de unidades de comportamiento de juveniles (Experieneia 3). — De 32 juveniles empleados, 24 hicieron tela (siete de ellos eran machos penultimos). Una tela se destruyo accidentalmente, previo a la observation. Diecinueve juveniles se VIERA Y COSTA— CAPTURA DE PRESAS POR MACHOS DE METEPEIRA 145 q uietud (11 ") de: mi aza to des plaza- mi ento 1 toqueteo toqueteo acica la acicala m iento miento mordedura prolongada q u i e t u d Cm n) transports man i pul a c ion J tensa- /T”. t ensa- miento CimciQX" " " > miento des mi( plaza e to desplaza miento acicala. acicala- miento miento Figs. 1-4. — Sucesiones de unidades de comportamiento observadas en la captura de hormigas ( Acromyrmex sp.) en cuatro grupos experimentales de Metepeira sp. A. Las unidades de comportamiento se sucedieron excluyendose entre si en el tiempo; las frecuencias de sucesion entre estas unidades se indicaron con el espesor de las flechas (1 mm- 10 sucesiones); las frecuencias iguales o menores al 10% del numero de observaciones no fueron dibujadas, a efectos de facilitar la comparacion visual: 1, diagrama de frecuencias de 19 hembras adultas en sus propias telas (Experiencia 1); 2, diagrama de frecuencias de 17 machos adultos ocupando telas de hembras adultas (Experiencia 2); 3, diagrama de frecuencias de 23 juveniles en sus propias telas (Experiencia 3); 4, diagrama de frecuencias de 19 machos adultos ocupando telas de juveniles (Experiencia 4). 146 THE JOURNAL OF ARACHNOLOGY observaron ioicialmente en el refugio y cuatro en el centre de la tela. Veintiun individuos iniciaron su actividad con desplazamiento y dos con tensamiento, Tensamiento se vinculo principalmente con desplazamiento y una unica vez con toqueteo. Toqueteo se relation© ademas con mordeduras cortas, envolvimiento, desplazamiento, transport© y quietud, Todos los individuos comenzaron la fase inmovilizacion con envolvimiento, que se vinculo estrechamente con mordeduras cortas, Ambas unidades se vincularon en menor medida con mordedura prolongada. Mordedura prolongada se relation© fundamentalmente con mordeduras cortas (Fig. 3). Catorce individuos comenzaron la fase terminal desde la unidad envolvimiento, dos despues de mordedura prolongada, dos despues de mordeduras cortas y uno despues de quietud. Dieciseis juveniles desarrollaron integramente la fase terminal, ingiriendo en el refugio. Quietud se vinculo principalmente con mordedura prolongada y envolvimiento, relacionandose con menor frecuencia con transport©, mordeduras cortas, toqueteo, desplazamiento y aeicalamiento, Seis individuos terminaron el comportamiento en quietud. Aeicalamiento fue realizado por tres individuos y se relation© principalmente con desplazamiento y en menor medida con envolvi- miento, quietud, transporte y manipulation, Un macho subadulto abandon© la presa despues de inmovilizarla y termino su comportamiento realizando aeicalamiento. No se observaron otras diferencias entre los machos subadultos y los otros juveniles. La eficiencia de captura de presas fue 100%. Sucesion de unidades de comportamiento de machos que utilizaron telas de juveniles (Experlenda 4). — De 23 machos empleados, 19 ocuparon telas: 14 se ubicaron en el refugio y cinco en el centro de la tela. Quince individuos realizaron inicialmente desplazamiento y cuatro comenzaron con tensamiento. Seis machos realizaron tensamiento y nueve toqueteo en la fase detection (Fig. 4). El pasaje de la fase detection a la fase inmovilizacion se realize con las sucesiones desplazamiento a envolvimiento o toqueteo a envolvimiento. La fase inmovilizacion present© tambien una fuerte relation entre envolvimiento y mordeduras cortas. Ambas unidades se vincularon en menor medida con mordedura prolongada (Fig. 4). La fase terminal fue realizada por 15 individuos: 11 a partir de envolvimiento y cuatro a partir de mordedura prolongada, Cuatro machos terminaron el comportamiento en quietud; uno de el los abandon© la presa una vez inmovilizada. Quietud se vinculo principalmente con envolvimiento y mordedura prolongada y en menor medida con toqueteo, desplazamiento, mordeduras cortas y transporte. Aeicalamiento se relation© fundamentalmente con mordeduras cortas, toqueteo y envolvimiento. No se observaron diferencias entre machos que ocuparon telas de machos subadultos y machos que ocuparon telas de otros juveniles. La eficiencia de captura de presas fue 95% (un solo macho no capture la presa). Comparacion entre los grupos experimentales. — Se compararon estadistica- mente las frecuencias absolutas de unidades de comportamiento entre: (i) Experiencia 1 y Experiencia 2; (ii) Experiencia 3 y Experiencia 4; (iii) Experiencia 1 y Experiencia 3; (iv) Experiencia 2 y Experiencia 4. Las unidades de VIERA Y COSTA— CAPTURA DE PRESAS POR MACHOS DE METEPEIRA 147 comportamiento cuyas frecuencias fueron menores al numero de observaciones (no preseetes en todas las observaciones), se compararon mediante el test de Fisher y las unidades de frecuencia mayor o igual al numero de observaciones (presentes en todas las observaciones) se compararon mediante el test de Wilcoxon. Las duraciones de las fases comportamentales se compararon mediante el test de Student. a. Comparaciones mediante el test de Fisher: Se compararon las frequencias de las unidades tensamiento, toqueteo, acicalamiento, transporte, manipulacion y quietud, como tambien la frecuencia de captura en los cuatro grupos experimentales (Tabla 1). La unidad tensamiento mostro diferencias significativas entre las experiencias 1 y 2, reflejando la mayor frecuencia presentada por los machos de la Experiencia 2. La unidad toqueteo mostro diferencias en dos comparaciones (Tabla 1), presentando frecencias altas en los dos grupos de machos (experiencias 2 y 4). Acicalamiento se observo con alta frecuencia tambien en machos, pero se observaron diferencias significativas solo en la comparacion Experiencia 1 -Experiencia 2. Las hembras mostraron una alta frecuencia en las unidades transporte y manipulacion (fase terminal), presentando diferencias significativas con los machos de la Experiencia 2. La baja frecuencia de estos ultimos determino tambien diferencias en la unidad transporte con los machos de la Experiencia 4. La unidad quietud no mostro' diferencias estadisticas entre los grupos, aunque se observo una frecuencia alta en los juveniles. La eficiencia de captura de presas fue maxima en los juveniles, alta en hembras y machos de la Experiencia 4 y baja en los machos de la Experiencia 2. Se diferencio estadisticarnente la frecuencia de captura de hembras con los machos de la Experiencia 2 (Tabla 1). b. Comparaciones mediante el test de Wilcoxon: Se compararon las frecuencias de las unidades desplazamiento, envolvimiento, mordeduras cortas y mordedura prolongada en los cuatro grupos experimentales (Tabla 2). No se encontraron diferencias estadisticarnente significativas en las frecuencias de las unidades desplazamiento y mordedura prolongada. Los machos de la Experiencia 2 presentaron frecuencias bajas en envolvimiento y mordeduras cortas, que resultaron distintas estadisticarnente respecto a hembras y machos de la Experiencia 4. c. Comparaciones generates de frecuencias: Los cuatro grupos experimentales presentaron un patron comun del comportamiento de captura sobre Acromyrmex sp., constituido por tres fases que se suceden en el tiempo. Los modelos de captura de hembras y juveniles fueron indistinguibles entre si a la luz de las comparaciones efectuadas, no presentando diferencias significativas en las frecuencias de unidades ni en la frecuencia de captura. Los juveniles se diferenciaron de los machos que capturaron en las mismas telas (Experiencia 4) solamente en la frecuencia de toqueteo. Los dos grupos de machos presentaron una fase deteccion mas compleja que hembras y juveniles, accediendo a la fase inmovilizacion no solo desde desplazamiento sino tambien desde toqueteo (Figs. 2 y 4). Los machos de la Experiencia 2 presentaron diferencias en las frecuencias de tres unidades de comportamiento respecto a los machos de la Experiencia 4, mientras que se diferenciaron en las frecuencias de siete unidades y en la captura de presas con las hembras (Tablas 1 y 2). Para tener una idea global comparativa de los cuatro grupos experimentales, se sometieron los valores de frecuencias de unidades a tecnicas de agrupamiento. La Fig. 5 permite destacar nuevamente la 148 THE JOURNAL OF ARACHNOLOGY Tabla 1. — Comparaciones entre las frecuencias de unidades de comportamiento y de captura de la presa en los cuatro grupos experimentales, utilizando el test de probabilidad exacta de Fisher (* = diferencias significativas). UNIDADES Exp. 1-Exp. 2 Exp. 3-Exp. 4 Exp. 1-Exp. 3 Exp. 2-Exp. 4 Tensamiento 0.016* 0.248 0.146 0.174 Toqueteo 6.150 X 10"4* 0.015* 0.301 0.209 Acicalamiento 0.028* 0.327 0.301 0.127 Transporte 0.016* 0.219 0.146 0.039* Manipulation 0.016* 0.222 0.146 0.075 Quietud 0.261 0.089 0.136 0.247 Captura 0.016* 0.452 0.198 0.060 disimilitud de los machos de la Experiencia 2 respecto a los otros grupos, que son muy similares entre si. d. Duracion de las fases comportamentales: — Los resultados obtenidos del registro temporal del comportamiento de los cuatro grupos experimentales se exponen en la Tabla 3. Se destaca una gran dispersion de todos los valores. Se compararon estos valores en los distintos grupos experimentales mediante el test de diferencia de medias de Student con restricciones para la varianza. Los resultados no mostraron diferencias estadisticamente significativas en las fases inmovilizacion y terminal, ni en la duracion total del comportamiento de captura; se observaron si diferencias en la fase deteccion en las hembras respecto a los machos de la Experiencia 2 y los juveniles (Tabla 4). Estos resultados reflejan la corta duracion de esta fase en las hembras. Los machos de la Experiencia 2 mostraron una fase deteccion sumamente extensa y a la vez la maxima variabilidad. e. Reparacion de la tela: — Aunque no se controlaron metodicamente todas las observaciones, se observe que hembras y juveniles repararon siempre la tela despues de la captura, mientras que los machos de las experiencias 2 y 4 no repararon nunca la tela. DISCUSION En el campo (Punta Espinillo) los autores observaron machos adultos de Metepeira sp. A ocupando telas orbiculares aparentemente coespecificas, capturando presas (cinco observaciones). En el laboratorio, los machos adultos de esta especie no construyen telas orbiculares, ocupan telas de juveniles y hembras adultas coespecificos y capturan presas, sin reparar las telas (Viera y Costa 1985). Tabla 2. — Comparaciones entre frecuencias de unidades de comportamiento en los cuatro grupos experimentales, utilizando el test de Wilcoxon (* = diferencias significativas). UNIDADES Exp. 1-Exp. 2 Exp. 3-Exp. 4 Exp. 1-Exp. 3 Exp. 2-Exp. 4 Desplazamiento P > 0. 1 P>0.1 P>0.1 P> 0.1 Envolvimiento 0.01 > P > 0.002* P> 0.1 P> 0.1 0.05 > P>0.01* Mordeduras cortas 0.05 > Z5 > 0.01* P> 0.1 P> 0.1 0.05 > P>0.01* Mordedura prolongada P> 0.1 7* > 0. 1 0.1 > P > 0.05 P> 0.1 VIERA Y COSTA— CAPTURA DE PRESAS POR MACHOS DE METEPEIRA 149 Fig. 5. — Fenograma de los cuatro grupos experi- mentales (OTUs) en base a las frecuencias absolu- tas de 10 unidades de comportamiento (caraeteres). Coefieiente de distancia de Crovello, UPGMA y tecnica de ligamiento promedio. HEMB = hembras en sus telas; MAHE = machos en telas de hembras; JUVE = juveniles en sus telas; MAJU = machos en telas de juveniles. Los resultados del preseete trabajo muestrao que los machos de esta especie maetieeen intacto el patron de captura de los estadios juveniles, inmovilizando las presas con seda y transportandolas al refugio, Tambien se observaron, en los machos, elementos motores particulares (cualitativa y cuantitativamente) proba- blemente vieeulados con el comportamiento sexual Los resultados de las experieecias de control (captura por hembras adultas y por juveniles) coincidieron basicamente con la description de Viera (1986). Sin embargo, deben destacarse algunas diferencias entre resultados anteriores (Viera 1986) y los del presente estudio, respecto a la predation de hembras y juveniles: (i) Viera describio uea fase de detection compleja (desplazamiento 1 — orientation — desplazamiento 2) que en el presente trabajo se observe simplificada (desplazamiento). Esta diferencia respond eri a a que en ei presente trabajo la presa se ubico en la vertical inferior de la tela y resulto inaparente la orientation de la at aha en el centro de la tela; (ii) Viera observe solo un caso de acicalamiento (4%) y aqui se observaron siete casos (17%); (hi) toqueteo se vinculo aqui con detection y en Viera con la fase inmovilizacion de la presa. Estas dos ultimas diferencias puedee ser debid as al azar o reflejar diferencias no controladas en el metodo, Las diferencias entre los trabajos senalan tambien la convenientia del uso metodico de grupos de control El comportamiento de captura realizado por hembras fue similar al de juveniles, excepto en la detection de la presa, que fue mas simple y r apid a en las hembras (Tabla 4). Esta caracteristica puede ser atribuida al aprendizaje y/o motivation alimeetaria, as oca ad a a la reproduction, mayores en las hembras. Con esta exception, los resultados obtenidos validan la metodo logia de Viera (1986) de reunir ambos grupos con fines descriptivos. En las experiencias 2 y 4 los machos presentaron una fase de detection compleja, con frecuencias altas de toqueteo y tensamiento. Tensamiento se observe con menor frecuencia en juveniles, aunque no se distinguio estadisticamente de los machos de la Experieecia 4. Toqueteo y tensamiento han sido observados por los autores en el cortejo de los machos de esta especie (datos 10 5 o * — - ■ i JUVE i MAJU HEMB MAHE Tabla 3. — Duraciones medias (X) y desviacion tipica (DT), en segundos, de las fases comportamentales desarrolladas por los cuatro grupos experimentales de Metepeira sp. A (N = numero de datos). Fases Experieecia 1 Experiencia 2 Experiencia 3 Experiencia 4 X ± DT N X ± DT N X ± DT N X ± DT N Detection 20.4+ 16.7 19 304.3+412.3 17 50.4+ 59.7 23 97.5+110.1 19 Inmovilizacion 839.9+614,8 19 714.8+482.9 12 1065.4+671.0 23 971.6+877.5 19 Terminal 75.9+110.2 16 72.8+ 75.4 8 207.0+255.3 17 86.0+ 79.6 15 Total 924.3+626.1 19 876.8+496.7 17 1268.8+728. S 23 1137.0+896.4 19 150 THE JOURNAL OF ARACHNOLOGY Tabla 4. — Comparacion entre las duraciones de las fases comportamentales desarolladas por los cuatro grupos experimentales, utilizando el test de Student (* = diferencias significativas). Fases Exp. 1-Exp. 2 Exp. 3-Exp. 4 Exp. 1-Exp. 3 Exp. 2-Exp. 4 Deteccion P= 0.014* P = 0.101 P = 0.028* P = 0.065 Inmovilizacion P = 0.562 P = 0.698 P = 0.266 P = 0.304 Terminal P = 0.940 P = 0.075 P = 0.063 P = 0.704 Total P = 0.803 P = 0.611 P = 0.108 P = 0.287 no publicados), lo que sugiere que elementos sexuales se intercalan frecuente- mente en el comportamiento alimentario de los machos. Es de destacar que tales elementos tambien se observaron en machos ubicados en telas de machos juveniles. Esto sugiere una fuerte motivacion sexual en los machos adultos, independiente de la presencia o no de estimulos tactoquimicos persistentes (feromona sexual de tela). La presencia de unidades sexuales en la fase inicial de captura contribuiria a la seguridad del macho, inhibiendo posibles ataques pradatorios de hembras en condiciones naturales. En los machos, acicalamiento se vinculo predominantemente con desplazamiento (Experiencia 2) o con mordeduras cortas (Experiencia 4): este resultado sugiere que esta unidad pueda cumplir funciones distintas en ambos casos (limpieza de sensores asociados a la deteccion de la presa y limpieza tegumentaria posterior a las mordeduras, respectivamente). Olivera-Curotti (1984), en Araneus suspicax (O.P.-Cambridge) y Viera (1986), en Meteperia sp. A, observaron acicalamiento vinculado a la ingestion de la presa. Los machos de la Experiencia 2 presentaron menores frecuencias absolutas de las unidades envolvimiento, mordeduras cortas y transporte, respecto a los machos de la Experiencia 4. Estas diferencias reflejan el hecho de que varios machos de la Experiencia 2 no realizaron fase inmovilizacion (no atacaron a la presa o no la retuvieron despues de demorarse en la deteccion) y consecuentemente no capturaron la presa. Las dificultades de los machos de la Experiencia 2 para inmovilizar probablemente respondan a una mayor interferencia de elementos sexuales en la fase deteccion que en los machos de la Experiencia 4, provocada por la feromona sexual persistente en el primer grupo. Las hembras y los machos de la Experiencia 2 presentaron entre si diferencias de frecuencias en siete unidades de comportamiento, en la frecuencia de captura de presas y en la duracion de la fase deteccion. Estas diferencias reflejan las caracteristicas inversas de ambos grupos: como ya fue dicho, las hembras presentaron una deteccion simple y frecuencias altas de unidades de las fases inmovilizacion y terminal; los machos, por el contrario, presentaron una deteccion muy compleja y frecuencias bajas de las unidades de comportamiento en las otras fases. Los juveniles y los machos de la Experiencia 4 mostraron comportamientos de captura semej antes entre si, exceptuando la frecuencia de toqueteo y las vinculaciones de acicalamiento; ambas caracteristicas fueron discutidas mas arriba. En terminos generales, los machos desarrollaron una secuencia completa de captura, orientandose inicialmente por las vibraciones de la presa y posterior- mente transportando la presa hasta el refugio. Resulta claro que, a diferencia del comportamiento constructor de telas, el comportamiento de captura persiste integramente en esta fase particular de la ontogenia. Salvo cinco machos VIERA Y COSTA— CAPTURA DE PRESAS POR MACHOS DE METEPEIRA 151 (Experiencia 2) que persistieron en las fase detection, los machos restantes mostraron completa habilidad de maniobra en un medio que ellos no construyeron. El alejamiento extreme de los machos de la Experiencia 2 en el feeograma de la Fig, 5 destaca la persistencia en la fase deteccion y tambien ref) eta el gran peso que la tecnica utilizada otorga a las frecuencias de sucesion mas altas (fase iemovilizacion). En Araneidae es frecuente que los machos compartan telas con hembras adultas o subadultas, compitiendo entre si para copular (Robinson y Robinson 1978, 1981; Christenson 1984; Vollrath 1980). Este tipo de competencia parece no existir en Meteperia sp. A, que presenta alta densidad poblacioeal y aparentemente una proportion equivalente de machos y de hembras. Los machos, de tamano similar a las hembras, no podrian compartir las telas orbiculares con estas por fuertes interferencias vibratorias y las propias exigencias alimentarias. Los machos de Meteperia sp. A mostraron la capacidad de alimentarse sin construir redes de captura ni interferir con las hembras. Esta tactic a parece adecuarse a una estrategia de machos poliginos y mas o menos longevos, Recientes observaciones de campo indican que los machos podrian recoeocer hembras subadultas y permanecer en los hiios perifericos a la tela. Resultan necesarias nuevas observaciones de campo sobre la dinamica de ocupacion de telas por machos ( 0.05) were Table 2. — Mean number of species occurring in the study sites on the ground surface. Collection Month COMMUNITY POND PINE x ± (SE) SAND PINE SCRUB 5c ± (SE) FLATWOODS 5c ± (SE) May 15.00 (1.16) 9.33 (1.34) 15.67 (2.03) July 13.67 (0.88) 14.33 (0.66) 10.00 (1.00) September 10.67 (1.20) 12.67 (0.34) 8.00 (2.09) November 8.00 (0.00) 8.00 (2.00) 7.00 (2.00) January 4.33 (0.66) 7.33 (1.34) 4.33 (0.88) March 10.33 (0.34) 10.33 (1.46) 9.67 (1.20) COREY AND TAYLOR— GROUND SURFACE SPIDERS IN FLORIDA 217 > Q 200 Fig. 3. — Seasonal distribution of the most common families of ground surface spiders caught in pitfall traps in pond pine (A), sand pine scrub (•), and flatwoods (■). Lycosidae (upper), Hahniidae (middle), and Salticidae (lower). 60 found between the mean number of species occurring in the three communities during the collecting months (Table 2). This result may be due to the large variance in number of species found among the communities. Species diversity, based on Simpson’s Index of Diversity (Simpson 1949), was low for all communities. Pond pine had a value of 0.71, sand pine scrub of 0.90, and flatwoods of 0.94. This might be due to the high species richness and small number of dominant species found. Spider families, represented by individuals collected on the ground surface, are listed in Table 3. Over all communities, the three most common families were lycosids, hahniids, salticids; collectively, they represent 72.5% of all spiders captured in pitfall traps. In pond pine, hahniids, lycosids, and ctenids represented 79.5% of that community’s total spider assemblage. In sand pine scrub, lycosids, salticids, and hahniids represented 80.3% of the total spider assemblage. In flatwoods, lycosids, hahniids, and salticids represented 70.6% of the total spider assemblage. Figure 3 shows seasonal abundance of three common families occurring on the ground surface. The species composition in our study differed from that found by Muma (1973) who studied ground surface spiders in sand pine dune and pine flatwoods near Winter Haven, Florida. Only seven species were common to his sand pine dune and our sand pine scrub habitats. These were Pholcomma hirsuta Emerton, Hahnia cinerea, Trochosa parthenus Simon, Sosippus floridanus Simon, Cesonia hilineata (Hentz), Drassyllus seminolus (Chamb. & Gertsch), and Castianeira floridana (Banks). In pine flatwoods, only Neoantista agilis (Keys.), Sosippus floridanus , and Oxyopes salticus (Hentz) were found in both studies. Reasons for the small number of spider species common to both studies are unknown. 218 THE JOURNAL OF ARACHNOLOGY Table 3.— Number of individuals collected and percent of spiders by family for the three communi- ties. POND PINE SAND PINE FLATWOODS TOTAL FAMILY # % # % # % # % Oecobiidae 3 0.3 0 0.0 1 0.3 4 0.2 Uloboridae 3 0.3 0 0.0 3 0.8 6 0.3 Dictynidae 2 0.2 0 0.0 0 0.0 2 0.1 Oonopidae 6 0.6 0 0.0 1 0.3 7 0.3 Pholcidae 2 0.2 0 0.0 1 0.3 3 0.1 Theridiidae 18 1.6 12 1.4 13 3.4 43 1.8 Mymenidae I 0.1 0 0.0 0 0.0 1 0.04 Linyphiidae 21 1.9 29 3.4 11 2.8 61 2.6 Linyphiinae 2 0.2 2 0.2 3 0.8 7 0.3 Erigoninae 19 1.7 26 3.1 8 2.1 53 2.3 Araneidae 1 0.1 1 0.1 0 0.0 2 0.1 Theridiosomatidae 5 0.5 0 0.0 0 0.0 5 0.1 Tetragnathidae 1 0.1 1 0.1 1 0.3 3 0.1 Agelenidae 2 0.2 4 0.5 1 0.3 7 0.3 Hahniidae 481 44.0 106 12.5 52 13.7 639 27.4 Lycosidae 344 31.4 424 49.8 180 47.2 948 40.8 Oxyopidae 7 0.6 4 0.5 2 0.5 13 0.6 Gnaphosidae 12 1.1 24 2.8 28 7.4 64 2.8 Clubionidae 26 2.4 16 1.9 14 3.7 56 2.4 Zoridae 1 0.1 0 0.0 0 0.0 1 0.04 Ctenidae 45 4.1 18 2.1 10 2.6 73 3.1 Sparassidae 0 0.0 0 0.0 1 0.3 1 0.04 Thomisidae 16 1.5 34 4.0 18 4.7 68 2.9 Salticidae 27 2.5 153 18.0 35 9.7 215 9.3 Undetermined 70 6.4 25 2.9 9 2.4 104 4.5 Our study and that of Muma’s (1973) show important differences in species compositions of spiders between communities. In our sand pine sites, lycosids, salticids, and hahniids comprised 80.3% of the spider population. In contrast, lycosids (53%), gnaphosids (19%), and salticids (18%) totaled 90% of the spider population in the sand pine dune studied by Muma (1973). In our flatwoods community lycosids, hahniids, and salticids comprised 70.6% of the total spider population, whereas 90% of the total population in Muma’s pine flatwoods consisted of lycosids (64%), salticids (21%), and linyphiids (5%). Differences in the two studies may be due to temporal changes in Florida habitats. Table 4 shows the 15 commonest species collected by frequency of occurrence. The three most common species for all communities were Hahnia cineria , Habrocestum bufoides Chamberlin & Ivie, and Pardosa sp. #1. Nineteen species occurred in all communities (Table 5). Hahnia cinerea was common in all communities and Sosippus floridanus and H. bufoides were common in sand pine scrub and flatwoods. Changes in the seasonal cycle were due to variation in the population of each individual species and also to the appearance and disappearance of species at different times of the year. The largest number of adult spiders in all three communities occurred during the summer. Three species had two different months with large population peaks; Centus captiosus Gertsch in July and January, Oxyptila modes ta (Scheffer) in November and March, and Zelotes pullus Bryant in September and March. Hahnia cinerea was present in large COREY AND TAYLOR— GROUND SURFACE SPIDERS IN FLORIDA 219 Table 4. — Fifteen spider species ranked by frequency of occurrence within each plant community. SPECIES POND PINE SAND PINE SCRUB FLATWOODS Hahnia cinerea Emerton 1 2 1 Schizocosa sp. 2 — Pardosa sp. #2 3 14 9 Ctenus captiosus Gertsch 4 9 8 Lycosa punctulata (Hentz) 5 10 Lycosa sp. #1 6 5 10 Schizocosa duplex Chamberlin 7 6 ■■ Ozyptila modesta (Scheffer) 8 7 5 Sosippus floridanus Simon 9 4 2 Pirata alachuus Gertsch & Wallace 10 — — Habrocestum bufoides Chamberlin & Ivie 11 1 3 Zelotes pullus Bryant 11 11 6 Thy modes sp. 11 — 12 Corythaiia sp. 11 — — Trachelas deceptus (Banks) 15 — — Trochosa partenus Simon Hi 8 7 Pardosa sp. #1 — 3 — Erigioninae sp. #3 — 10 — Theridion alabamense Gertsch & Wallace 13 — Neoantistea agilis (Key.) sKiP^ 4 Lycosidae sp. #2 — 1 1 — Lycosidae sp. #3 — 14 — Castianeira floridana (Banks) — 14 — Drassyllus sp. — — 12 Lycosidae sp. #1 — — 12 Litophyllus temporarius Chamberlin — — 15 Table 5. — Spiders found in all three plant communities and their relative abundance (R = rare, less than 1% of the total population for that community; P = present, 1-4.9%; and C = common, 5% or more). SPECIES POND PINE SAND PINE SCRUB FLATWOODS Pholcomma hirsutum Emerton R R R Theridion alabamense Gertsch & Wallace R R P Thymoites sp. R R R Erigoninae sp. #3 R P R Hahnia cinerea Emerton C C C Schizocosa duplex Chamberlin R P P Schizocosa sp. C R R Sosippus floridanus Simon P C C Pardosa sp. #1 R c P Pardosa sp. #2 C R P Lycosa sp. #1 P P P Zelotes pullus Bryant R P P Litopyllus temporarius Chamberlin R R P Castianeira floridana (Banks) R R R C. longipalpus (Hentz) R R R Ctenus captiosus Gertsch P P P Oxyptila modesta (Scheffer) P P P Habrocestum bufoides Chamberlin & Ivie R C C Metacyrba sp. R R R 220 THE JOURNAL OF ARACHNOLOGY numbers from July through September in pond pine and flatwoods, but in January in sand pine scrub. Ctenus captiosus appeared in large numbers in summer in pond pine and sand pine scrub, but in the fall in fid i woods. Berry (1971) found that adults and juveniles of some species appeared in large numbers after a period of time when no or very few adults or juveniles were found. Sosippus floridanus, Trochosa parthenus, Zelotes pullus , and Habroces - turn bufoides exhibited this behavior in our study. These species were found in small numbers in November through March and in large numbers beginning in May. These species may overwinter as juveniles or eggs. A new species of Drassyllus was found in ffatwoods (Platnick, pres. comm.). Four males and one female were caught in May at sites G (three individuals) and I (two individuals). One female Zora pumila (Hentz) was found in May at site B of the pond pine community; the previous southernmost limit of its range was Alabama (Kaston 1978). ACKNOWLEDGMENTS We thank Jonathan Reiskind, Jonathan Coddington, G. B. Edwards, Norman I. Platnick, and James Redner for examining specimens. We thank I. Jack Stout, David H. Vickers, and Henry O. Whittier for their comments and assistance. A1 Fliss helped construct the pitfall traps and Mike Albig assisted in the field collections. Allen R. Brady and Daniel T. Jennings made helpful comments on an earlier draft of the manuscript. This work was supported by a grant to David T. Corey from the Exline-Frizzeli Fund for Arachnological Research, grant No. 8. LITERATURE CITED Barnes, R. D. 1953. The ecological distribution of spiders in non-forest maritime communities at Beaufort, North Carolina. Ecol. Monog., 23:315-337. Berry, J. W. 1970. Spiders of the North Carolina Piedmont old-field communities. J. Elisha Mitchell Sci. Soc., 86:97-105. Berry, J. W. 1971. Seasonal distribution of common spiders in the North Carolina Piedmont. American Mid. Nat., 85(2):526-531. Bultman, T. L. and G. W. Uetz. 1982. Abundance and community structure of forest floor spiders following litter manipulation. Oecologia, 55:34-41. Bultman, T. L., G. W. Uetz and A. R. Brady. 1982. A comparison of cursorial spider communities along a successional gradient. J. ArachnoL, 10:23-33. Cady, A. B. 1984. Microhabitat selection and locomotor activity of Schizocosa ocreaia (Walckenaer) (Araneae: Lycosidae). J. ArachnoL, 11:297-307. Corey, D. T. 1987. Arachnid fauna in three central Florida plant communities. M.S. thesis, University of Central Florida, Orlando, FL. 290 pp. Corey, D. T. and W. K. Taylor. 1987. Scorpion, pseudoscorpion, and opilionid faunas in three central Florida plant communities. Florida Scient., 50(3): 162-167. Duffey, E. 1962. A population of spiders in Limestone Grassland. J. Animal Ecol, 31:571-599. Gertsch, W. J. 1979. American Spiders. Van Nostrand Reinhold Co., New York, 274 pp. Kaston, B. J. 1978. How to Know the Spiders. Wm. C. Brown Co. Pub., Dubuque, Iowa. 272 pp. Krebs, C. J. 1978. Ecology: The Experimental Analysis of Distribution and Abundance. Harper & Row, Publishers, Inc., New York. 678 pp. Lowrie, D. C. 1942. The ecology of the spiders of the xeric dunelands of the Chicago area. Bull. Chicago Acad. Sci., 6(9): 161-189. Lowrie, D. C. 1948. The ecological succession of spiders of the Chicago area dunes. Ecology, 28(3):334-351. COREY AND TAYLOR— GROUND SURFACE SPIDERS IN FLORIDA 221 Lowrie, D. C. 1963. The effects of grazing and intensive collecting on a population of the green lynx spider. Ecology, 44(8):777-781. Lowrie, D. C. 1968. The spiders of the herbaceous stratum of the Jackson Hole region of Wyoming. Northwest Sci., 42(3):89-100. Lowrie, D. C. 1971. Effects of time of day and weather on spider catches with a sweep net. Ecology, 52(2):348-351. Lowrie, D. C. 1985. Preliminary survey of wandering spiders of a mixed coniferous forest. J. Arachnol., 13:97-110. Muma, M. H. 1973. Comparison of ground surface spiders in four Central Florida ecosystems. The Florida Entomoh, 56(3): 173-196. Rey, J. R. and E. D. McCoy. 1983. Terrestrial arthropods of Northwest Florida salt marshes: Araneae and Pseudoscorpiones (Arachnida). The Florida Entomol., 66(4):497-503. Simpson, E. H. 1949. Measurement of diversity. Nature, 163:688. Simpson, G. G., A. Roe and R. C. Lewontin. 1960. Quantitative Zoology. Harcourt, Brace and Co., New York. 440 pp. Turnbull, A. L. 1960. The spider population of a stand of oak ( Quercus robur L.) in Wytham Wood, Berks., England. Canadian Entomoh, 92:110-124. Uetz, G. W. 1975. Temporal and spatial variation in species diversity of wandering spiders (Araneae) in deciduous forest litter. Environ. Entomoh, 4(5):719-724. Uetz, G. W. 1977. Coexistence in a guild of wandering spiders. J. Animal Ecology, 46:531-541. Uetz, G. W. 1979. The influence of variation in litter habitats on spider communities. Oecologia (Berh), 40:29-42. Wunderlin, R. P. 1982. Guide to the Vascular Plants of Central Florida. University Press of Florida, Tampa, Florida. 472 pp. Manuscript received September 1987, revised February 1988. Jennings, D. T. and J. B. Dimond. 1988. Arboreal spiders (Araneae) on balsam fir and spruces in East-Central Maine. J. Arachnol., 16:223-235. ARBOREAL SPIDERS (ARANEAE) ON BALSAM FIR AND SPRUCES IN EAST-CENTRAL MAINE Daniel T. Jennings Northeastern Forest Experiment Station USDA Building, University of Maine Orono, Maine 04469 USA and John B. Dimond Department of Entomology University of Maine Orono, Maine 04469 USA ABSTRACT Spiders of 1 1 families, 22 genera, and at least 33 species were collected from crown foliage samples of Abies balsamea (L.) Mill, Picea rubens Sarg., and Picea glauca (Moench) Voss in east-central Maine. For both study years (1985, 1986), spider species composition varied by foraging strategy (web spinner, hunter) and among 10 study sites. Numbers, life stages, and sex ratios of spiders also differed between study years. Spider densities per in of foliage area generally were greater ( P ^ 0.05) on spruces ( X - 16.3 ±1.1) than on fir (X = 10.9 + 1.0). Estimates of absolute populations of arboreal spiders ranged from 35,139 to 323,080/ha; of spruce budworm from 271,401 to 6,122,919/ha. Spider- budworm densities/ha covaried significantly (P ^ 0.001) each year (r = 0.84, 1985; r = 0.71, 1986). None of the measured forest-stand parameters (basal area, tree species percentage) were reliable predictors of spider populations/ ha. INTRODUCTION Recent devastating outbreaks of the spruce budworm, Choristoneura fumife- rana (Clem.), have renewed interest in determining potential natural enemies of this defoliator of northeastern spruce-fir forests. Because of their ubiquitous occurrence, relative abundance, and predatory habits, spiders are considered important predators of the spruce budworm (Morris 1963; Jennings and Crawford 1985). Watt (1963) estimated that only 0.49 larvae/ m2 of tree foliage would have to be eaten by predators, including spiders, to account for a decrease in population survival rate of the spruce budworm. A necessary first step for determining potential natural enemies of any pest is identification of the associated fauna. Some information is available about spiders of northeastern spruce-fir forests; however, most studies concern the terricolous fauna (Freitag et al. 1969; Carter and Brown 1973; Varty and Carter 1974; Jennings et al. 1988; Hilburn and Jennings 1988). Few previous studies have dealt with the arboreal spiders found on foliage of spruces ( Picea spp.) and firs {Abies spp.); only one study (Jennings and Collins 1987) was completed in Maine. 224 THE JOURNAL OF ARACHNOLOGY During recent investigations of microsporidia-infected budworms, arboreal spider and spruce bud worm populations were assessed on balsam fir, Abies balsamea (L.) Mill., red spruce, Picea rubens Sarg., and white spruce, Picea glauca (Moench) Voss trees in east-central Maine. This paper describes the arboreal-spider fauna associated with these coniferous tree species, compares spider-spruce budworm population densities among study sites and between host- tree species, estimates absolute population densities of spiders and budworms per hectare, and explores possible relationships among spiders, budworms, and forest- stand parameters. METHODS Study areas. — Six forest stands were investigated in 1985; four were investigated in 1986. All study sites were in east-central Maine (Fig. 1), and were in open, fir-spruce stands that had moderate to heavy infestations of spruce budworm. Site locations, abbreviations, and sampling years were: Big Lake (BL)-T27 ED, Washington County; 1985. Deer Lake (DL-’85)-T34 MD, east, Hancock County; 1985. Deer Lake (DL-’86)-T34 MD, south, Hancock County; 1986. Eastern Road (ER)-Upper Molunkus Twp., Aroostook County; 1985. Machias River (MR)-T31 MD, Washington County; 1986. Myra (MY-’85)-T32 MD, Hancock County; 1985. Myra (MY-,86)-T32 MD, east, Hancock County; 1986. Old Stream (OSVT31 MD, Washington County; 1986. Raven (RA)-Macwahoc Pit., Aroostook County; 1985. River Road (RR)-Mattawamkeag Twp., Penobscot County; 1985. At each study location, linear transects (0.5 to 1 km) were established along old logging roads or forest trails. We used a variable-size plot design to facilitate tree selection. Branch samples were obtained with a long, extendable pole pruner. Ten or 20 dominant/ codominant trees of each category (balsam fir-red spruce; balsam fir-white spruce; balsam fir) were selected, flagged, and numbered for consecutive sampling on a weekly basis. Stand measurements. — The variable-plot sample method (Wenger 1984) was used to determine basal areas (m2/ha) of balsam fir, spruces (both red and white), northern white-cedar, Thuja occidentalis L. , eastern white pine, Pinus strobus L. , eastern hemlock, Tsuga canadensis (L.) Carr., and hardwoods (mostly Acer spp. and Betula spp.). At each study site (except DL-?85), ten 10-factor prism plots were established and all trees 2.54 cm tallied by species or species group (e.g., spruces, hardwoods). Only four prism plots were taken at DL-’85. Branch samples. — Trees were sampled at about weekly intervals both study years. In 1985, sampling began 20 May and ended 12 July; in 1986, sampling began 10 June and ended 2 July. Each year, the duration of the sampling period corresponded with the early larval (L3-L4) through pupal stages of the spruce budworm. This allowed determination of predator-prey densities when budworm larvae and pupae are susceptible to predation (Morris 1963). At each sampling, one 45-cm branch was pruned from the upper crown half of each sample tree. Sectional, aluminum pole primers equipped with a cloth-basket attachment below the pruner head were used to cut and lower branches to the JENNINGS AND DIMOND— ARBOREAL SPIDERS EAST-CENTRAL MAINE 225 Fig. 1. — Study-site locations in east-central Maine, 1985, 1986. (See text for abbreviations). ground. Jennings and Collins (1987) found that more spiders were collected when pole pruners were equipped with a catchment basket than with a clamping device (Stein 1969). At ground level, the severed branch and any dislodged arthropods were removed from the basket and placed in a labeled plastic bag for transport to the laboratory. In the laboratory, trained technicians clipped branches into small lengths (8 to 10 cm) and closely searched all foliage for spruce budworms and spiders. All collected spiders were stored in 2-dram vials containing 75% ethanol. Spider identifications. — Only sexually mature spiders were identified to species, following identification keys and species descriptions of Kaston (1981) and other consulted sources. Juveniles, including penultimate stages, were identified to generic level, and juveniles of two recognizable species groups ( Philodromus aureolus and P. rufus) were assigned to either group based on color patterns of legs, carapace, and abdomen (Dondale and Redner 1978). Representative specimens of all identified species are deposited in the arachnid collection, U.S. National Museum of Natural History, Washington, DC. Data analyses. — Branch surface areas were calculated by the formula: A - (L X W)/2, where L is branch length and W is maximum width (Sanders 1980). Population densities were expressed as spiders or spruce budworms/ m2 of branch foliage area. To estimate absolute populations, we converted spider-budworm densities/ m2 of foliage area to densities/ ha by the method of Morris (1955). Populations were computed as: N spiders/ ha = [X spiders/ m2 of fir foliage • ^ BSAF + 226 THE JOURNAL OF ARACHNOLOGY N X spiders/ m2 of spruce foliage • ^ BSASp], where Sp = l N ^ BSAf = sum of branch surface areas of N fir trees/ ha; F=1 N ^ BSA5p = sum of branch surface areas of N spruce trees/ ha. Sp = 1 The following equations were used to calculate branch surface area per tree based on diameter at breast height (dbh): BSA F = — 6.93 + 3.43 dbhcm, after Morris (1955) BSA^ = 2.64 + 3.34 dbhcm, after Dimond (unpubl.) Nonparametric procedures (Sokal and Rohlf 1981) were used for most statistical tests at P = 0.05. The Kruskal-Wallis Test (SAS Institute 1985) was used to compare spider-budworm densities among study sites and between tree species. Pearson correlation coefficient was used to test the interdependence of spider-budworm densities. Regression analyses were used to explore possible relationships between spider populations and measured stand parameters. RESULTS Forest stands. — Percentage composition of tree species by basal area (m2/ha) indicated that most study sites had softwood components of balsam fir and spruces (mostly red spruce) (Table 1), with occasional eastern white pine, eastern hemlock, and northern white-cedar. With few exceptions, hardwoods accounted for < 30% of total basal area. Deer Lake (DL-’85) had a relatively high percentage of eastern hemlock; Myra (MY-’86) and Raven (RA) had high percentages of balsam fir. Mean basal areas of fir and spruces generally were < 10 m2/ha (Table 2), characteristic of open-grown stands. Mean tree diameters of firs ranged from 9.1 ± 3.9 cm to 17,5 ± 2.2 cm; mean diameters of spruces ranged from 10.7 ± 0.5 cm to 29.6 ± 2.3 cm. Tree heights of dominant/codominant sample trees were 10 to 15 m. Spider taxa. — Spiders of 1 1 families, 22 genera, and at least 33 species were collected from arboreal habitats of spruce-fir forests in east-central Maine (Table 3). Fewer families, genera, and species were collected in 1986 than in 1985, but not unexpectedly because only balsam fir was sampled in 1986, and the sampling period was shorter. For both study years, species composition of spiders differed by foraging strategy; species of web spinners were slightly more prevalent in branch samples (56.2% of total species, 1985; 58.8% of total species, 1986) than species of hunters (43.8%, 1985; 41.2%, 1986). Species per family ranged from one (Linyphiidae, Oxyopidae) to six (Erigonidae). In 1985, equal numbers of species (25) were collected from foliage JENNINGS AND DIMOND— ARBOREAL SPIDERS EAST-CENTRAL MAINE 227 Table L — Percentage species composition of forest stands investigated for spiders and spruce bud worms, east-central Maine, 1985-86. Site Fir Spruces Pine Hemlock Cedar Hardwoods 1985 Study Sites BL 27.4 15.9 15.0 29.2 12.4 DL-’85 11.1 27.8 9.3 50.0 1.8 ER 4.3 20.5 39.3 25.6 10.2 MY- ’85 40.0 21.1 2.2 1.1 35.6 RA 53.7 3.7 2.2 3.0 37.3 RR 33.6 10.9 29.1 0.9 25.5 1986 Study Sites DL-586 44.4 41.6 13.9 MR 35.3 22.2 3.0 13.1 7.1 19.2 MY-’86 90.9 4.6 2.3 2.3 OS 45.6 14.7 7.4 4.4 27.9 samples of balsam fir and red spruce; only 12 species were collected from foliage of white spruce. In 1986, 17 species were collected from balsam fir foliage, the only tree species sampled that year. Because sampling intensities differed between years, adult spiders of 15 species were collected in 1985 but not in 1986; whereas, adults of only two species, Mangora placida (Hentz) and Eris militaris (Hentz), were collected in 1986 but not in 1985. Adults of 14 species were collected in both years. Composition of spider species differed among study sites each year, no doubt because the sites were not identical (Tables 1 and 2) both years. In 1985, adult species per site ranged from 8 to 20 (X = 14.3); in 1986, from 6 to 14 (X = 10.0). Species common to all six sites sampled in 1985 were Dictyna brevitarsus Emerton, Theridion murarium Emerton, Pityohyphantes costatus (Hentz), Grammonota angusta Dondale, and Metaphidippus flaviceps Kaston. Both Grammonota pictilis (G.P.-Gambridge) and Philodromus exilis Banks were found on five study sites in 1985. Only one species, Metaphidippus : flaviceps , was common to all four sites sampled in 1986; however, Theridion differens Emerton, Pityohyphantes costatus , and Grammonota angusta were each found on three sites. Table 2. — Mean (+SE) basal areas of balsam' fir and spruces in forest stands investigated for spiders and spruce bud worms, east-central Maine, 1985-86. Mean (+SE) basal area (m2/ha). Site ■ Fir ■ ■-•Spruces 1985 Study Sites BL 7.1 (2.4). 4.1 (2.2) OL-585 3.4 (2.0) 8.6 (2.0) ER 1.2 (0.6) 5.5 (1.0) MY-’85 8.3 (3.1) 4.4 (1.0) RA 16.5 (2.3) 1.2 (0.5) RR 8.5 (1.7) 2.8 (1.0) 1986 Study Sites DL-’86 3.7 (0.5) 3.4 (0.8) MR 8.0 (1.0) 5.0 (1.4) MY- '86 9.2 (1-3) 0.5 (0.3) OS 7.1 (0.9) 2.3 (1.0) 228 THE JOURNAL OF ARACHNOLOGY Table 3. — Arboreal spiders on foliage of Abies balsamea, Picea rubens, and Picea glauca , east- central Maine, 1985-86. FAMILY Species BALSAM FIR SPRUCES 5 $ juv. 6 $ juv. WEB SPINNERS DICTYNIDAE Dictyna brevitarsus Emerton 6 13 5 15 Dictyna phylax Gertsch & Ivie 2 4 6 Dictyna sp. 24 39 THERIDIIDAE Theridion differens Emerton 1 Theridion montanum Emerton 3 4 Theridion murarium Emerton 5 7 4 10 Theridion sp. 37 39 LINYPHIIDAE Pityohyphantes costatus (Hentz) 1 10 3 8 Pityohyphantes sp. 7 ERIGONIDAE Ceraticelus atriceps (O.P.-Cambridge) 1 Ceraticelus sp. 1 Dismodicus bifrons decemoculatus (Emerton) 1 Grammonota angusta Dondale 10 56 17 59 Grammonota pictilis (O.P.-Cambridge) 4 10 4 11 Grammonota sp. 5 Pocadicnemis americana Millidge 1 1 Walckenaeria lepida (Kulczynski) 1 ARANEIDAE Araniella displicata (Hentz) 3 9 5 Araniella sp. 1 3 Araneus sp. (nr. saevus) 1 1 Araneus sp. 1 Cyclosa conica (Pallas) 1 Cyclosa sp. 1 Mangora placida (Hentz) 3 Neoscona arabesca (Walckenaer) 1 Neoscona sp. 4 2 TETRAGNATHIDAE Tetragnatha versicolor Walckenaer 1 2 Tetragnatha viridis Walckenaer 1 1 Tetragnatha sp. 2 4 Subtotals 32 119 70 37 121 103 HUNTERS OXYOPIDAE Oxyopes sp. 1 CLUBIONIDAE Clubiona canadensis Emerton 2 1 Clubiona trivialis C. L. Koch 1 1 Clubiona sp. 4 11 PHILODROMIDAE Philodromus exilis Banks 1 6 2 6 Philodromus pernix Blackwall 1 3 Philodromus placidus Banks 5 6 Philodromus praelustris Keyserling 1 Philodromus rufus vibrans Dondale 1 Philodromus sp. ( aureolus grp.) 3 1 JENNINGS AND DIMOND— ARBOREAL SPIDERS EAST-CENTRAL MAINE 229 Philodromus sp. ( rufus grp.) Philodromus sp. THOMISIDAE Misumena vatia (Clerck) Misumena sp. Xysticus punctatus Keyserling Xysticus sp. SALTICIDAE 14 26 15 33 1 4 2 19 1 40 Eris militaris (Hentz) Eris sp. Metaphidippus flaviceps Kaston Metaphidippus protervus (Walckenaer) Metaphidippus sp. Salticus scenicus (L.) 6 23 1 47 6 1 13 64 Subtotals 10 46 104 12 30 176 TOTALS 42 165 174 49 151 279 Spider numbers, life stages, sex ratios. — Over half (62.6%) of the total collected spiders (n - 765) in 1985 were from spruces. This high percentage was unexpected because of the distribution of branch samples among tree species in 1985, i.e., balsam fir (n = 60), red spruce (n = 50), and white spruce (n - 10). In 1986, all of the collected spiders (n - 95) were from balsam fir trees (n = 80). In 1985, most collected spiders were juveniles (53.6%), followed by females (35.7%), and males (10.7%). In 1986, both juveniles and females were equally abundant (45.3% each) among collections, with fewer males (9.5%). Sex ratios of males to females were 1:3.3 in 1985 and 1:4.8 in 1986. Spider densities. — For both study years, spider populations/ m2 of foliage area varied among study sites (Tables 4 and 5, column 2fs). In 1985, most sites had comparable means of 10 to 14 spiders/ m2 of balsam fir foliage and 19 to 20 spiders/ m2 of spruce foliage. In 1986, most sites had means of 7 to 14 spiders/ m2 of balsam fir foliage; spruce was not sampled that year. For unknown reasons, some sites had significantly fewer spiders (balsam fir-RA, DL-’86; spruces-ER, RA) than other sites. Spider densities/ m2 of foliage generally were greater on spruces than on balsam fir (Table 4, row 2fs); these differences were significantly greater ( P ^ 0.05) on three of the study sites and over all sites in 1985. However, guild densities by foraging strategy (web spinner, hunter) were not significantly different ( P ^ 0.05) among tree species in 1985. Budworm densities. — Populations of spruce bud worm larvae and pupae/ m2 of foliage also varied among study sites both years (Tables 4 and 5, column Xs). In 1985, mean densities were about equal between host tree species for most sites and over all sites. In 1986, most study sites had mean densities > 200 budworms/ m2 of balsam fir foliage; all sites X - 224.4 (± 10.4). Absolute populations. — Estimates of arboreal spiders/ ha ranged from 80,745 (± 17,643) to 323,080 (± 114,839) in 1985 [X = 192,073 (± 28,171)]; from 35,139 (± 5,338) to 287,024 (± 40,853) [X = 175,047 (± 20,287)] in 1986. Some of the observed variation among sites may be attributed to differences in percentage species composition of balsam fir and spruces (Table 1); however, spider densities/ ha were weakly correlated with percentage fir (r = —0.08, 1985; r = 0.03, 1986), but more closely with percentage spruce (r = 0.30, 1985). Differences in percentages of balsam fir and spruces profoundly affected estimates of spiders/ ha 230 THE JOURNAL OF ARACHNOLOGY Table 4. — Densities of spiders and spruce bud worms/ m2 of foliage, balsam fir, red and white spruces, east-central Maine, 1985. *= White spruce sampled on MY-’85; red spruce sampled on all other sites. Column means (ab) followed by the same letter(s) are not significantly different, SAS Institute (1985), Kruskal-Wallis Test, P = 0.05. Row means (xy) followed by the same letter(s) are not significantly different, SAS Institute (1985), Kruskal-Wallis Test, P- 0.05. SPIDERS X(± SE)/m2 SPRUCE BUDWORMS *(+SE)/m2 1985 SITES Balsam fir Spruces* Balsam fir Spruces* BL 13.9ax (2.5) 20.2ay (2.7) 146. lbx (17.2) 99.4cy (11.5) DL-’85 10.6ax (1.8) 20.4ay (3.2) 167. 7bx (22.6) 179. 4bx (28.0) ER 10. Sax (3.8) 9.6bx (1.9) 54.8cx (11.9) 25.8dx ( 4.2) MY-’85 12.7ax (2.4) 19.0ay (2.6) 212.6ax (21.9) 217. 6ax (21.5) RA 3.3bx (0.8) 8.4bx (1.8) 35.2cx ( 4.5) 34.3dx ( 6.3) RR 14.3ax (2.4) I8.9ax (2.9) 145. 6bx (17.5) 130.7bcx (14.6) ALL 10. 9x (1.0) 16. 3y (1.1) 129.9x ( 7.8) 1 17. 8x ( 7.9) between tree species on the same site. For example, although spiders/ m2 of foliage were not significantly different between sampled tree species for study site ER in 1985 (Table 4), significantly more (. P Si 0.03) spiders/ ha were estimated to occur on spruces than on balsam fir, largely due to the preponderance of spruces (5X > fir) on this site. The same pattern of influence also was evident for balsam fir; however, when host-tree differences were Si 2X, then spider densities/ ha tended to equalize between tree species. Estimates of absolute populations of spruce budworms/ha ranged from 271,401 (± 67,590) to 3,076,290 (± 928,941) in 1985 [X = 1,821,159 (± 273,181)]; from 2,465,473 (± 347,069) to 6,122,919 (± 1,091,369) in 1986 [X = 4,258,870 (± 422,723)]. Spider-budworm relationships. — Correlation analyses indicated positive signifi- cant associations between spider and budworm densities/ ha each study year (Figs. 2 and 3). Spider and budworm densities covaried slightly more together in 1985 (r = 0.84, P is 0.001) than in 1986 (r - 0.71, P is 0.001), which might be attributed to population estimates derived from only one tree species in 1986. The scattered data points at relatively high spider-budworm densities (i.e., ^ 400,000 spiders, is 4.5 million budworms) indicated greater variation above these densities in 1985 (Fig. 2) than in 1986 (Fig. 3). Spider/forest stands. — Regression analyses indicated that none of the measured forest-stand parameters were reliable predictors of spider populations/ ha (Table 6). For unknown reasons, total basal area, fir basal area, and percent spruce were better indicators (i.e., higher r2 values) of spider populations in 1986 than in 1985. Table 5. — Densities of spiders and spruce budworms/ m2 foliage, balsam fir, east-central Maine, 1986. Column means (ab) followed by the same letter are not significantly different, SAS Institute (1985), Kruskal-Wallis Test, P = 0.05. 1986 SITES SPIDERS X (+SE)/m2 SPRUCE BUDWORMS X(±SE)/m2 DL-86 3.2b (1.0) 289.9a (20.7) MR 9.4a (2.2) 118.6b (11.5) MY-’86 7.0a (1.6) 235.2a (17.5) OS 13.9a (2.4) 255.8a (24.6) ALL 8.5 (1.0) 224.4 (10.4) JENNINGS AND DIMOND— ARBOREAL SPIDERS EAST-CENTRAL MAINE 231 Fig. 2. — Association of spider-budworm densities per hectare, six study sites, east-central Maine, 1985. (Pearson Correlation Coefficient, r = 0.84, P Si 0.001). Small, medium, and large circles represent one, two, and three observations, respectively. DISCUSSION Spider taxa. — Many of the species of spiders we collected on foliage of balsam fir, red and white spruces in east-central Maine have been taken on balsam fir in New Brunswick (Loughton et al. 1963; Renault 1968) and on red spruce in northern Maine (Jennings and Collins 1987). Species not previously recorded from arboreal habitats of Maine’s spruce-fir forests include Pocadicnemis americana Millidge, Walckenaeria lepida (Kulczynski), Tetragnaiha viridis Walckenaer, Philo dr omus praelustris Keyserling, Eris militaris (Hentz), and Metaphidippus flaviceps Kaston. Based on frequency of collection (this study and Loughton et al. 1963; Renault 1968; Renault and Miller 1972; Jennings and Collins 1987), we consider the following species as typical arboreal spiders of northeastern spruce-fir forests: Dictyna brevitarsus Emerton, D. phylax Gertsch & Ivie, Theridion montanum Emerton, T. murarium Emerton, Pityohyphantes costatus (Hentz), Ceraticelus atriceps (O.R-Cambridge), Grammonota angusta Dondale, Araniella displicata (Hentz), Clubiona canadensis Emerton, C. trivialis C. L. Koch, Philodromus exilis Banks, P. placidus Banks, and Xysticus punctatus Keyserling. Our observed differences in composition of spider species by foraging strategy (web spinners, 55%; hunters, 45%) generally agree with earlier studies. Jennings and Collins (1987) collected more species of web spinners (54%) than hunters (46%) from red spruce foliage in northern Maine (n - 21 species). Likewise, Jennings and Hilburn (1988) captured more species of web spinners (56%) than hunters (44%) in Malaise traps operated in spruce-fir forests of west-central Maine (n = 25 species). Even greater percentages of web spinners (67%) than hunters (33%) were reported associated with balsam fir foliage in New Brunswick, where n - 54 species (Loughton et al. 1963). We conclude that the web-spinner guild comprises a major species component of the arboreal spider fauna associated with northeastern spruce-fir forests. The observed dissimilarities in composition of spider species among study sites may be within the realm of expected variation for northeastern spruce-fir forests. 232 THE JOURNAL OF ARACHNOLOGY Fig. 3. — Association of spider-bud worm densities/ ha, four study sites, east-central Maine, 1986. (Pearson Correlation Coefficient, r - 0.71, P 0.001). Small, medium, and large circles represent one, two, and three observations, respectively. Renault and Miller (1972:1045-46) noted “a remarkable constancy in the species composition in any one location,” but “a marked dissimilarity in the species composition in different areas,” all classed as fir-spruce biotype. Until additional studies are completed, the overall species composition of arboreal spruce-fir spiders remains undefined except for localized areas. No doubt, additional species will be added to the known fauna by the use of other sampling methods (e.g., whole-tree counts), extension of sampling periods, and increased sample sizes. Spider numbers, life stages, sex ratios. — Some of the observed differences in spider numbers, life stages, and sex ratios can be attributed to reproductive cycles of individual species and production of young spiderlings during midsummer. At least five of the species found during this study, Theridion murarium Emerton, A r amelia displicata (Hentz), Misumena vatia (Clerck), Xysticus punctatus Keyserling, and Philodromus placidus, have biennial life histories (Dondale 1961, 1977). Juveniles of Theridion, Xysticus , and Philodromus commonly were collected among branch samples, especially in 1985, and probably represented immature stages of biennial species. The biased sex ratio in favor of females was not unexpected because female spiders generally live longer than males (Gertsch 1979), and males generally are more vagrant and hence less likely to be sampled than females. However, we are unable to explain why female spiders were slightly more prevalent among collections in 1986 (45.3%) than in 1985 (35.7%). Sample size (i.e., 878 branches in 1985 vs. only 298 branches in 1986) and sampling-time differences between years may have been contributing factors. Spider densities. — The spider densities/ m2 of foliage that we observed in east- central Maine generally are greater than those previously reported elsewhere. Morris (1963) noted densities of 2.65 spiders/ 10 ft2 (2.85/m2) of balsam fir foliage in June and 2.34 spiders/ 10 ft2 (2.52/m2) in July, on the Green River Watershed, New Brunswick. For two plots on the same watershed and sampling dates comparable to our Maine study (22 May to 12 July), Loughton et al. (1963) reported densities ranging from 0.6 spiders/ 10 ft2 (0.6/m2) to 18.9 spiders/ 10 ft2 (20.3/ m2) of balsam fir foliage. We calculated mean densities for these same plots and sampling periods as: K2 = 9.5 spiders/m2 (1957), 7.9 spiders/m2 (1958); G16 = JENNINGS AND DIMOND— ARBOREAL SPIDERS EAST-CENTRAL MAINE 233 Table 6. — Coefficients of determination (r2) for predicting spider populations/ ha based on forest stand parameters. STAND PARAMETER 1985 1986 r2 P 2 r P Total basal area 0.06 0.65 0.31 0.44 Fir basal area 0.03 0.73 0.48 0.31 Spruce basal area 0.08 0.59 0.04 0.81 Percent fir 0.01 0.88 0.00 0.97 Percent spruce 0.09 0.56 0.56 0.25 7.9 spiders/ m2 (1957), 5.6 spiders/m2 (1958). Renault and Miller (1972) noted a yearly constancy of about 8.2 spiders/ m2 of balsam fir foliage for a 9-year period (1962-70), on the Green River Watershed, New Brunswick. For spiders on spruces, Morris (1963) reported densities of 4.8 spiders/ 10 ft2 (5.2/m2) in June and 12.5 spiders/ 10 ft2 (13.4/m2) in July, Green River Watershed, New Brunswick. In northern Maine, Jennings and Collins (1987) observed densities ranging from 1.5 to 16.6 spiders/ m2 and a mean density of 7.1 spiders/ m2 of red spruce foliage sampled in late July. The overall mean density of 16.3 spiders/m2 of spruce foliage in east-central Maine was substantially greater than expected based on previous studies. Absolute populations. — Our estimates of absolute populations of arboreal spiders/ha generally were less than some earlier findings, i.e., Morris (1963) estimated 187,500 spiders/ ha in New Brunswick; Haynes and Sisojevic (1966) estimated 3 12,500/ ha in New Brunswick; Jennings and Collins (1987) estimated 645,853/ ha in north-central Maine. We suspect that stand species composition and stand density are important determinants of absolute populations of arboreal spiders. For example, the sites investigated in east-central Maine were open grown, fir-spruce stands; whereas, those estimated to harbor 645,853 spiders/ ha were dense, red spruce stands (Jennings and Collins 1987). No doubt our estimates and those earlier are conservative because not all represented tree species were sampled. Total absolute populations of spiders/ ha are apt to be much higher when estimates include all tree species and all strata (arboreal, epigeal, terrestrial). Spider-budworm relationships. — Based on the observed high correlations between spider and budworm densities, we suspect that spiders were responding to available prey (budworm) populations in east-central Maine. All life stages of the spruce budworm — eggs, larvae, pupae, and adults — are susceptible to spider predation, but the larvae are particularly vulnerable because of their relative abundance, size, and activity. Loughton et al. (1963) investigated spider predation on the spruce budworm in New Brunswick and concluded that: (1) erigonids are the most important predatory group because of their large numbers; (2) theridiids are the most effective predators, based on percentages showing positive feedings on budworm; and (3) salticids are important predators at all stages of budworm larval development, including the late instars. Predation on the large larvae is especially important because mortality during the late larval stage influences generation survival of the spruce budworm (Morris 1963). All three spider families (Erigonidae, Theridiidae, Salticidae) were common among branch samples from balsam fir and spruces in east-central Maine. 234 THE JOURNAL OF ARACHNOLOGY Spiders/forest stands. — Small sample sizes (n = 6 sites, 1985; n - 4 sites, 1986) may have contributed to the weak relationships between forest-stand parameters and estimates of spider populations/ ha in east-central Maine. Because of greater spider densities and lower coefficients of variation on spruce, we predict that percentage spruce will be a more reliable indicator of absolute spider populations/ ha than percentage fir. However, numerous other factors, such as canopy closure, crown class, and stand age, warrant investigation. Jennings and Collins (1987) hypothesized that red spruce may harbor more spiders, both individuals and species, than balsam fir. Our results in east-central Maine partially support this hypothesis, Le., overall, significantly more (P is 0.001) spiders/ m2 of foliage were found on spruces (red, white) than on balsam fir. We also found equal numbers of spider species (25) despite unequal sample sizes in favor of balsam fir. We suspect that additional spider species may occur on each tree species and that spruces provide greater microhabitat space for web building and for foraging by hunting spiders. In Minnesota, Stratton et al. (1979) found that white spruce had more spider individuals and species than red pine, Pinus resinosa Ait., and northern white-cedar. They attributed the greater spider diversity on spruce to differences in plant physiognomy, Le., structure of needles and branches. Results of our study in east-central Maine indicate that: (1) host-tree species influences arboreal-spider density per m2 of foliage area, (2) percentage composition of tree species in forest stands affects overall estimates of arboreal- spider populations/ ha, and (3) estimates of spider-budworm (predator-prey) densities may be highly correlated. Additional studies are needed to define and evaluate other factors that possibly influence spider populations on northeastern conifers. If indeed spruce supports greater populations of spiders than fir, forest pest managers will have the option to manage forests to increase populations of potential predators of the spruce bud worm. Such options offer alternatives to reliance solely on chemical insecticides. ACKNOWLEDGMENTS We thank Bruce A. Watt and Terri L. Preston for technical assistance. Special thanks are due C. D. Dondale and J. H Red tier, Biosystematics Research Centre, Ottawa, for identifications of some Erigonidae. Richard A. Hosmer provided computer programming assistance; Janet J. Melvin provided word processing service. We thank our' reviewers, C. D, Dondale, D. H. Morse, and G. E. Stratton, for their constructive comments on an earlier draft. D. W. Seegrist, Northeastern Forest Experiment Station, Broomall, PA, provided statistical review. LITERATURE CITED Carter, N. E. and N. R. Brown. 1973. Seasonal abundance of certain soil arthropods in a fenitrothion- treated red spruce stand. Canadian Entomoh, 105:1065-1073. Dondale, C. D. 1961. Life histories of some common spiders from trees and shrubs in Nova Scotia. Canadian J. Zool, 39:777-787. Dondale, C. D. 1977. Life histories and distribution patterns of hunting spiders (Araneida) in an Ontario meadow. J. Arachnoh, 4:73-93. JENNINGS AND DIMOND— ARBOREAL SPIDERS EAST-CENTRAL MAINE 235 Dondale, C. D. and J. H. Redner. 1978. The crab spiders of Canada and Alaska (Araneae: Philodromidae and Thomisidae). Canadian Dep. Agric. Publ., 1663. 255 pp. Freitag, R., G. W. Ozburn, and R. E. Leech. 1969. The effects of sumithion and phosphamidon on populations of five carabid beetles and the spider Trochosa terricola in northwestern Ontario and including a list of collected species of carabid beetles and spiders. Canadian Entomol., 101:1328- 1333. Gertsch, W. J. 1979. American Spiders. 2nd. ed. Van Nostrand Reinhold Co., New York. 272 pp. Haynes, D. L. and P. Sisojevic. 1966. Predatory behavior of Philodromus rufus Walckenaer (Araneae: Thomisidae). Canadian Entomol., 98:113-133. Hilburn, D. J. and D. T. Jennings. 1988. Terricolous spiders (Araneae) of insecticide-treated spruce-fir forests in west-central Maine. The Great Lakes Entomol. (in press). Jennings, D. T. and J. A. Collins. 1987. Spiders on red spruce foliage in northern Maine. J. Arachnol., 14:303-314. Jennings, D. T. and H. S. Crawford, Jr. 1985. Predators of the Spruce Budworm. U. S. Dep. Agric., Agric. Handb. 644. 77 pp. Jennings, D. T. and D. J. Hilburn. 1988. Spiders (Araneae) captured in Malaise traps in spruce-fir forests of west-central Maine. J. Arachnol., 16: (in press). Jennings, D. T., M. W. Houseweart, C. D. Dondale and J. H. Redner. 1988. Spiders (Araneae) associated with strip-clearcut and dense spruce-fir forests of Maine. J. Arachnol., 16: (in press). Kaston, B. J. 1981. Spiders of Connecticut. 2nd. ed. Bull Connecticut State Geol. Nat. Hist. Surv., 70. 1020 pp. Loughton, B. G., C. Derry and A. S. West. 1963. Spiders and the spruce budworm. Pp. 249-268, In The Dynamics of Epidemic Spruce Budworm Populations. (R. F. Morris, ed.). Mem. Entomol. Soc. Canada 31. 332 pp. Morris, R. F. 1955. The development of sampling techniques for forest insect defoliators, with particular reference to the spruce budworm. Canadian J. Zool., 33:225-294. Morris, R. F., (ed). 1963. The Dynamics of Epidemic Spruce Budworm Populations. Mem. Entomol. Soc. Canada 31. 332 pp. Renault, T. R. 1968. An illustrated key to arboreal spiders (Araneida) in the fir-spruce forests of New Brunswick. Canada Dep. Fisheries and Forestry, For. Res. Lab., Fredericton, New Brunswick. Int. Rep., M-39. 41 pp. Renault, T. R. and C. A. Miller. 1972. Spiders in a fir-spruce biotype: abundance, diversity, and influence of spruce budworm densities. Canadian J. Zool., 50:1039-1046. Sanders, C. J. 1980. A summary of current techniques used for sampling spruce budworm populations and estimating defoliation in eastern Canada. Canadian For. Serv., Great Lakes For. Res. Cent. Rep., O-X-306. 33 pp. SAS Institute. 1985. User’s Guide: Statistics, Version 5 Edition. SAS Institute, Cary, N.C. 956 pp. Sokal, R. R. and F. J. Rohlf. 1981. Biometry. 2nd. ed. W. H. Freeman Co., New York. 859 pp. Stein, J. D. 1969. Modified tree pruner for twig sampling. USDA For. Serv., Rocky Mount. For. Range Exp. Stn. Res. Note, RM-130. 2 pp. Stratton, G. E., G. W. Uetz and D. G. Dillery. 1979. A comparison of the spiders of three coniferous tree species. J. Aranchol., 6:219-226. Varty, I. W. and N. E. Carter. 1974. Inventory of litter arthropods and airborne insects in fir-spruce stands treated with insecticides. Canadian For. Serv., Maritimes For. Res. Cent. Inf. Rep., M-X- 48. 32 pp. Watt, K. E. F. 1963. The analysis of the survival of large larvae in the unsprayed area. Pp. 52-63, In The Dynamics of Epidemic Spruce Budworm Populations. (R. F. Morris, ed.). Mem. Entomol. Soc. Canada 31. 332 pp. Wenger, K. F. 1984. Forestry Handbook. 2nd. ed. John Wiley & Sons, New York. 1335 pp. Manuscript received December 1987, revised March 1988. Murphree, C. S. 1988. Morphology of the dorsal integument of ten opilionid species (Arachnida, Opiliones). J. Arachnol., 16:237-252. MORPHOLOGY OF THE DORSAL INTEGUMENT OF TEN OPILIONID SPECIES (ARACHNIDA, OPILIONES) C. Steven Murphree Department of Biology Middle Tennessee State University Murfreesboro, Tennessee 37132 USA ABSTRACT Specimens of Siro exilis Hoffman, Vonones sayi (Simon), Erebomaster sp., Leiobunum vittatum (Say), Leiobunum holtae McGhee, Hadrobunus maculosus (Wood), Eumesosoma nigrum (Say), Odiellus pictus (Wood), Caddo agilis Banks, and Hesperonemastoma kepharti (Crosby and Bishop) were investigated using scanning electron microscopy. Features of the dorsal integument of each specimen are described using available terminology. Variations in the generalized tuberculate-granulate morphology were observed in eight of the ten species studied. V sayi, C. agilis , and O. pictus exhibit a morphological gradient in features of their dorsal integuments. The presence of micropores is reported from the apices of tubercles of L. vittatum, L. holtae, H. maculosus , and O. pictus and from the cuticular backgrounds of S. exilis, V sayi, and E. nigrum. The morphological descriptions and comparisons presented provide a terminology for describing opilionid cuticular features. INTRODUCTION Species of the order Opiliones are often characterized by prominent protuberances, spines, and ornamented cuticles, especially in the suborder Laniatores (Shear and Gruber 1983). Studies with scanning electron microscopy (SEM) reveal still other, smaller, cuticular features. In the past, morphological studies of arthropod integuments have been primarily limited to insects. Although various morphological descriptions of the arachnid cuticle exist in the literature, a consistent descriptive terminology is currently unavailable. Previous light microscopical studies of arachnid integuments include those of Edgar (1963), Immel (1964), Grainge and Pearson (1966), Kennaugh (1968), Dalingwater (1975, 1981, 1987), and van der Hammen (1985), among others. Selected SEM studies of arachnid taxa exclusive of the Opiliones include those of Brody (1970), Pittard and Mitchell (1972), Woolley (1974), Crowe (1975), Quintero (1975), Platnick and Gertsch (1976), Platnick and Shadab (1976), Mutvei (1977), Keirans and Clifford (1978), Hill (1979), Hadley and Filshie (1979), Emerit (1981), Hadley (1981), Opell (1983), Platnick (1986), and Igelmund (1987). SEM studies of various opilionid species were conducted by Juberthie and Massoud (1976), Martens (1979), Martens and Schawaller (1977), and Spicer (1987). Martens (1978) referred to both macro- and micromorphological features 'Taken in part from an M.S. Thesis completed at Middle Tennessee State University, Murfreesboro, Tennessee. 2Current address: Department of Entomology, Auburn University, Alabama 36849-5413. 238 THE JOURNAL OF ARACHNOLOGY of the opilionid integument. Shear (1986) and Shear and Gruber (1983) used SEM to illustrate characters for cladistic analysis of ischyropsalidoid and troguloid opilionids. The purpose of this paper is to illustrate the cuticular features of ten opilionid species, propose a descriptive terminology for these features, and make preliminary comparisons of related species based on their cuticular features. METHODS AND MATERIALS Sources of material. — Specimens were obtained from the private collection of Dr. Charles R. McGhee at Middle Tennessee State University and from localities in Rutherford, Bedford, Cannon, McMinn, and Hickman counties of Tennessee. One species, Eumesosoma nigrum (Say), was on loan from the American Museum of Natural History. Method of study. — All specimens were preserved in 70% ethanol prior to preparation for SEM. The specimens were allowed to air dry or were freeze-dried from quick-frozen distilled water using a Thermovac lyophilizer. Freeze-drying was used for those species whose cuticles became distorted with air desiccation. The dried specimens were mounted on aluminum stubs using an epoxy glue for larger specimens and double-stick adhesive tape for smaller specimens, coated with gold-palladium in a Technics Hummer VI sputtering system, and examined at an accelerating voltage of 15 kV in an ISI SX-30 SEM. Photomicrographs were made with a Pentax MX 35 mm camera using Kodak™ Plus-X film. The surface features illustrated by the photomicrographs were compared with existing micrographs and morphological descriptions of both arachnid and insect integuments in an attempt to apply an appropriate descriptive terminology. The investigations of Harris (1979), Byers and Hinks (1973), Hinton (1970), and Kennaugh (1968) as well as the general works of Torre-Bueno (1962), Steinmann and Zombori (1981), and Baker and Wharton (1952) were used as primary sources of descriptive terminology. Magnification and terminology. — The photomicrographs are reproduced at magnifications which best illustrate the proposed terminology. Measurements in micrometers (um) are given to provide size comparisons. Because many descriptive terms exist with both diminutive and superlative forms, micromorphological descriptions require standards for selecting proper terms. The author agrees with Shear and Gruber (1983) who used the prefix “micro-” preceding terms describing features which measure 0.01 mm or less in size. Often, a species’ integument is sufficiently detailed to warrant the use of two, rarely three, descriptive components. This procedure is favored over the arbitrary creation of terms in a field often burdened with excessive synonymy. RESULTS The features of the dorsal integument of ten opilionid species representing nine genera and six families within the three suborders of the Opiliones were examined and described (Table 1). No cuticular differences attributable to sexual dimorphism were observed in species for which both sexes were available for study. A systematic list of the species studied is given in Appendix 1. Cuticular MURPHREE— OPILIONID CUTICULAR MORPHOLOGY 239 Table 1. — A summary of the cuticular features of 10 opilionid species. Refer to Appendix 1 for higher taxonomy. S. exilis V. sayi £. sp. L. vitt. L. holt. H. mac. E. nig. 0. pic. C. agilis H. kepharti Tuberculate X X X X X X X Convex Rounded X X X X X Basally constricted Microgranulate Laminate X X X X X X Spirally Perpendicularly X X X X Glabrous Microtuberculate X X X X Convex X Oblong Obtuse/ subdeltoid Reticulate X X X Microgranulate Denticles Micropores X X X X X X On tubercles Not on tubercles X X X X X X X Foveolate X X X Cycloid facetodea X X X X X Microgranulate X X X X X X X X Two-ranked Dentate X X X X X Imbricate X X X Subimbricate X X Laminate X X Acute/ obtuse X X Keeled X X Mucronate X Reticulate Rivulose X X X Rugose-plicate Sinuate X X Striate X Rectilinear setae Arcuate setae X X X X X X X X X X Substriate setae X X X X X X X Spirally X X X X X X X Setae on areolae X X Torose areolae Setae on microareolae X X X X X X X Rounded Depressed X X X X X X Setae on microalveolae X comparisons between related species (see Discussion) are made to better distinguish the characters of each species and are not intended to show phylogenetic relationships. Applied descriptive terminology is defined in Appendix 2. Siro exilis Hoffman. — Dorsal integument: A tuberculate-microgranulate mor- phology is observed from both the cephalothoracic and abdominal tergites of this species (Fig. 1). The convex tubercles are smooth above with abbreviated 240 THE JOURNAL OF ARACHNOLOGY Figs. 1-4 — Opilionid dorsal integument morphology: 1-2, S. exilis ; L dermal gland micropore (DM); 2, abdominal seta; 3-4, V sayi , 3, anterior cephalothorax; 4, posterior cephalothorax with dermal micropores. Scale = 10 /am, except Fig. 3, 100 /mi. microstriae surrounding their bases. The microgranulations of the cuticular background are regularly spaced, two-ranked, and appear as rounded points. The larger microgranulations measure 0.5 /.tin in basal diameter and as many as 40 are present in a given 25 /art2. The smaller microgranulations measure 0.25 /.tin in basal diameter and as many as 70 are represented in 25 //nr. The openings of dermal gland micropores (Juberthie and Massoud 1976) are seen as microfoveolae (Fig. 1). Each opening is encircled by 15-17 of the larger microgranulations. Abdominal setae: Rectilinear- aci cular setae arise from depressed microalveolae (Fig. 2). Vonones sayi (Simon). — Dorsal integument: The anterior cephalothoracic region exhibits a tuberculate-rivulose-microgranulate morphology (Fig. 3). The tubercles are located anterior to the ocularium and appear as rounded microgranular elevations of the cuticle. No micropores are visible on the surfaces of the tubercles. The cuticular surface has a subimbricate background of polygonal plates which exhibit a pattern of microgranular, sinuate furrows. The furrows are in relief of non-parallel ridges formed by fusion of the microgranulations. Numerous micropores appear as foveolae and are encircled by microgranulations. The microgranulations vary considerably in size and shape, with a maximum basal diameter of 0.4 /am. Both the posterior cephalothoracic and anterior abdominal regions exhibit a rivulose-microgranulate morphology (Fig. 4) which closely resembles that of the anterior cephalothorax, with the exception that no tubercles or polygonal plates are seen. The openings of micropores are also seen in this region. MURPHREE— OPILIONID CUTICULAR MORPHOLOGY 241 Figs. 5-8 — Opilionid dorsal integument morphology: 5-6, V. sayi ; 5, posterior abdomen; 6, cephalothoracic seta; 7-8, Erebomaster sp.; 7, dorsum; 8, abdominal seta. Scale = 50 yum (Figs. 5-6), 10 yum (Figs. 7-8). The posterior abdominal region of V sayi is imbricate-rivulose-microgranulate (Fig. 5). The imbrications or laminae are essentially similar to the polygonal plates of the anterior cephalothoracic region. The laminae are, however, keeled distally and range from obtuse to acute. Micropores also are seen in this region. Cephalothoracic setae: Arcuate-acicular setae lie parallel to the surface of the integument (Fig. 6). The setae are spirally substriate with basal, torose areolae which are rivulose-microgranulate. Setae are found only in the lateral areas of the cephalothorax, in certain areas of the posterior abdomen, and on the appendages. Erebomaster sp. — Dorsal integument: A microtuberculate-rivulose-microgranu- late morphology is observed from both the cephalothoracic and abdominal tergites of this species (Fig. 7). The numerous, convex microtubercles are superficially reticulate in some cases while others exhibit a punctulate morphology. The microgranulations of the cuticular background are of variable size and shape and have a maximum basal diameter of 0.1 gm. Abbreviated asymmetrical ridges are formed by fusion of the microgranulations. Micropores, if present, can not be distinguished from either the punctulations of the microtubercles or the coalescent microgranular background. Abdominal setae: Rectilinear, thick-shafted setae lie parallel to the surface of the integument (Fig. 8). The setae have prominent basal areolae which, like the microtubercles, are rivulose-microgranulate. Leiobunum vittatum (Say). — Dorsal integument: A tuberculate-microgranulate morphology is observed from both the cephalothoracic and abdominal tergites of this species (Fig. 9). The tubercles appear as rounded elevations of the cuticle and 242 THE JOURNAL OF ARACHNOLOGY Figs. 9-12. — Opilionid dorsal integument morphology: 9-10, L. vittaturrr, 9, micropore (M); 10, abdominal setae; 11-12, L. holtae ; 11, micropore; 12, abdominal seta. Scale = 50 /am (Figs. 10, 12), 100 /am (Figs. 9, 1 1). are covered by a number of laminae which form a loosely aggregated spiral. A micropore is visible at the summit of each tubercle. The microgranulations are subequal in size and appear as dentate projections of the integument. The basal width of the microgranulations ranges from 5 to 8 jum. The less sclerotized, transverse and lateral areas of the dorsum are microgranulate and devoid of tubercles. Abdominal setae: Thick-shafted, arcuate-acicular setae arise from rounded microareolae which do not resemble the larger tubercles or the smaller microgranulations (Fig. 10). The setae are spirally substriate. Leiobunum holtae McGhee.— Dorsal integument: As illustrated in Fig. 11, both the cephalothoracic and abominal tergites of this species are tuberculate- microgranulate. The cuticular morphology closely resembles that of L. vittatum. All regions of the dorsal integument of L. holtae exhibit both tubercles and microgranulations. Abdominal setae: As illustrated in Fig. 12, the setae of L. holtae closely resemble those of L. vittatum . Hadrobunus maculosus (Wood). — Dorsal integument: A tuberculate-microgran- ulate morphology is exhibited by both the cephalothoracic and abdominal tergites of this species (Fig. 13). The cuticular morphology closely resembles that of the two Leiobunum species with the exception of smooth, pointed denticles which are sparsely distributed. The denticles measure approximately 50 jum in height and 40 pm in basal width. Tubercles, denticles, and micrograeulatioes are observed from all regions of the dorsal integument. Numerous circular structures consisting of MURPHREE— OPILIONID CUTICULAR MORPHOLOGY 243 Figs. 13-16. — Opilionid dorsal integument morphology: 13-15, H. maculosus\ 13, micropore (M) and denticle (DT); 14, cycloid facetodea; 15, abdominal seta; 16, E. nigrum. Scale = 100 g.m, except Fig. 15, 10 Mm. many closely associated papillae range from 50 to 65 /im in diameter and appear in transverse rows across the dorsum (Fig. 14). For the time being, I term these structures “cycloid facetodea” since they resemble the multifaceted compound eyes of insects. Abdominal setae: As illustrated in Fig. 15, the setae of H. maculosus closely resemble those of the two Leiobunum species. Eumesosoma nigrum (Say). — Dorsal integument: A pronounced tuberculate- microgranulate morphology is observed from both the cephalothoracic and abdominal tergites of this species (Fig. 16). The prominent tubercles are subspherical and constricted basally. The tubercles show radial symmetry above with abbreviated striae radiating from a rounded process (Fig. 17). The tubercles measure approximately 40 ^m in diameter and 30 pm in height and no micropores are distinguishable above. The microgranulations are of variable size and appear as dentate projections of the integument. The basal width of the microgranulations ranges from 5 to 8 pm. Sparsely distributed micropores appear as microfoveolae and are partially encircled by incomplete microareolae (Fig. 18). The microareolae measure approximately 5 pm in diameter. Abdominal setae: Arcuate-acicular, thick-shafted setae arise from concave depressions atop rounded microareolae (Fig. 19). The microareolae do not resemble the other cuticular features and range from 7 to 9 pm in diameter. The setae are spirally substriate. Odiellus pictus (Wood). — Dorsal integument: The cephalothoracic region of O. pictus exhibits a tuberculate-imbricate-reticulate morphology (Fig. 20). The 244 THE JOURNAL OF ARACHNOLOGY Figs. 17-19. — Opilionid dorsal integument morphology: E. nigrum\ 17, tubercle; 18, dermal micropores (DM); 19, abdominal setae. Scale = 10 jum. tubercles appear as rounded elevations of the integument, each possessing a micropore at its apex. Most of the tubercles are glabrous above but some are encircled by oblong laminae which are perpendicular to the cuticular surface. The imbrications are distributed as laminae and form a shingle-like arrangement. The distal margins of the laminae range from obtuse to acute. The reticulations appear as marginal elevations of the laminae. A tuberculate-imbricate morphology is observed from the anterior abdominal region of this species (Fig. 21). The tubercles closely resemble those of the cephalothoracic region. However, the laminae are strongly keeled distally and occur in thickset layers. The distal margins of the laminae are acute. The posterior abdominal region is imbricate-mucronate (Fig. 22). Tubercles are absent from the last two abdominal segments. The laminae occur in thickset layers and each ends in a fine point or mucro. Abdominal setae : Two types of arcuate-acicular, thick-shafted setae are observed from the abdominal region. One form arises from external processes which are much larger than the abdominal tubercles, measuring approximately 40 pm in height and 35 pm in basal diameter (Fig. 23). The other type of setae arise from smaller tubercles measuring approximately 12 pm in both height and basal diameter (Fig. 22). Each of these seta-bearing tubercles is located within an incomplete alveolus and is subparallel to the surface of the integument. Caddo agilis Banks. — Dorsal integument: A subimbricate-microtuberculate morphology is observed from the cephalothoracic region, particularly the ocularium of this species (Fig. 24). The surface of the integument has a subimbricate background of polygonal plates and microtubercles of variable size. MURPHREE— OPILIONID CUTICULAR MORPHOLOGY 245 Figs. 20-23. — Opilionid dorsal integument morphology: O. pictus\ 20, cephalothorax; 21, anterior abdomen showing micropore (M); 22, posterior abdomen and seta; 23, abdominal seta. Scale = 50 jam, except Fig. 20, 100 pm. The larger microtubercles range from obtuse to subdeltoid and line the distal margins of the imbrications. Numerous smaller, obtuse microtubercles cover each of the polygonal imbrications. No micropores are observed from any region of the dorsum. The abdominal region of C. agilis is rugose-plicate-microtuberculate (Fig. 25). The microstructure of the integument resembles that of the cephalothoracic region with the exception that the polygonal imbrications are not present. The cuticular surface exhibits a folded pattern of impressed striae in relief of torose plications. The microtubercles closely resemble those of the cephalothoracic region and project from the plications at irregular intervals. The integument of C. agilis is more easily distorted when desiccated prior to SEM than that of the other species examined. Abdominal setae: Thick-shafted, arcuate-acicular setae arise from rectangular depressions atop microareolae (Fig. 26). The mcroareolae range from 7 to 9 gm in basal diameter and do not resemble other cuticular features. The setae are spirally substriate. Hesperonemastoma kepharti (Crosby and Bishop). — Dorsal integument: A microtuberculate-microgranulate morphology is observed from both the cephalo- thoracic and abdominal tergites of this species (Fig. 27). The numerous oblong, convex microtubercles exhibit microgranulations above and are constricted basally. A concave depression surrounds the posterior one-half of each microtubercle. The posterior margins of the microtubercles extend over the concave depressions and range from obtuse to acute. In the posterior abdominal 246 THE JOURNAL OF ARACHNOLOGY Figs. 24-26. — Opilionid dorsal integument morphology: C. agilis; 24, cephalothorax; 25, abdomen; 26, abdominal seta. Scale = 10/um. region, some of the posterior tubercular margins exhibit a trident of acute microdenticles (Fig. 28). The microgranulations are of variable size and are often sparsely distributed upon primarily glabrous areas of the integument. No micropores are distinguishable from either the surfaces of the microtubercles or the microgranular background. Abdominal setae : The setae are arcuate-acicular, spirally substriate, and thick- shafted (Fig. 28). They arise from the posterior margins of rounded microareolae at approximately a 45 degree angle to the integument. DISCUSSION Juberthie and Massoud (1976) conducted an SEM study of six cyphophthalmid species exclusive of S. exilis and reported that the species examined have similar cuticular features. The cuticular morphology of S. exilis closely resembles that of the six species studied by Juberthie and Massoud. Variations in cuticular morphology among the Sironidae include differences in the ratios of the two sizes of microgranulations per unit area, variations in the position of dermal gland micropores (e.g., a number of micropores are adjacent to the tubercles in Metasiro americanus (Davis)), and differences in the shape and size of the tubercles (Juberthie and Massoud 1976). This indicates that seven of the more than 50 known cyphophthalmid species have the same basic cuticular morphology. Although Shear (1980) did not use cuticular morphology as a taxonomic character in his reclassification of the Cyphophthalmi, further studies of representatives of the related family Pettalidae and superfamilies Stylocelloidea MURPHREE— OPILIONID CUTICULAR MORPHOLOGY 247 Figs. 27-28. — Opilionid dorsal integument morphology: H. kepharti ; 27, dorsum; 28, abdominal seta. Scale = 10 fim. and Ogoveoidea are needed to determine the importance of integumental microstructure in classifying cyphophthalmids. The cuticular morphology of V sayi most closely resembles that of Erehomaster sp.? although these species are only distantly related. A rivulose- microgranulate cuticular microstructure and a horizontal setal position are exhibited by both species. However, Erehomaster sp. exhibits neither an imbricate morphology nor any apparent micropores. The tubercles of V sayi differ in shape and number from those of Erehomaster sp. A morphological gradient exists in the cuticular features of V sayi which is illustrated by the progression from cephalothoracic tubercles to abdominal imbrications. Eisner et al. (1971) reported leg dabbing as a defensive measure of V. sayi. This may indicate that the cuticle of this species does not function in dispersing defensive secretions since these chemicals are actively applied to the potential predator rather than diffusely disseminated over the dorsal integument to form a “chemical shield”. A chemical shield is produced by the laniatorid Stygnomma spinifera (Packard) whose repellent secretions flow along lateral grooves of its dorsum, possibly by capillary action, before spreading over the dorsum (Duffield et al. 1981). The author has observed similar lateral grooves in an undetermined species of the Phalangodidae. No lateral grooves were observed on Erehomaster sp. in the present study. Although no live material was available for observations of the defensive behavior of Erehomaster sp., ventral grooves below the scent glands and hirsute segments of tarsi I are present. These leg dabbing characters are similar to those illustrated by Eisner et al. (1977) for two neotropical species of the Cosmetidae. The cuticular morphology of L. vittatum to a great extent resembles that of its congener, L. holtae , and the confamilial H. maculosus. The dorsum of L. vittatum exhibits transverse and lateral bands which are devoid of tubercles and which were not observed in the other opilionids examined. Martens (1978) made reference to lateral “Kanalen” or channels on the surface of the opilionid integument which may function in dispering the secretion of the scent glands. The transverse and lateral bands observed in L. vittatum may function in dispersing its copious scent gland secretion. The size and morphology of the setae of L. vittatum are with few exceptions similar to those of L. holtae, H. maculosus, E. nigrum, C. agilis , and H. kepharti. Spicer (1987) illustrated a type of palpal mechanoreceptor (sensilla chaetica) from Leiobunum townsendi Weed that 248 THE JOURNAL OF ARACHNOLOGY exhibits “whorled striae” Apparently, both sensory and non-sensory setae exhibit the spirally substriate morphology in Leiobunum , although additional species should be examined. Only the absence of microgranulate transverse and lateral bands distinguishes L holtae from L. vittatum in terms of cuticular morphology. Also, only the absence of denticles and cycloid facetodea may be used to distinguish L, vittatum and L holtae from H. maculosus. Although various types of cycloid facetodea were observed in V sayi , Erebomaster sp., O pictus , and C. agilis, only those of H. maculosus are illustrated since they distinguish this species from L, vittatum and L. holtae in terms of cuticular microstructure. The function of these possibly glandular structures is unknown and histological studies are needed. The cuticular morphology of E. nigrum is interesting because of the characteristic form of the tubercles. Cokendolpher (1980) referred to “obtuse tubercles scattered over the entire dorsum” in descriptions of the six species of Eumesosoma. The tubercular micropores observed in L. vittatum , L. holtae , H. maculosus , and O. pictus are not present in the confamilial E. nigrum. The setae of E. nigrum are similar in both size and structure to those of L. vittatum , L. holtae , H. maculosus, C. agilis , and H. kepharti but differ in that they are emitted from concave depressions atop the microareolae. The microstructure of the integument of O. pictus is distinct because of its prominent laminar imbrications which cover the dorsum. The diversity in form of the laminae from the different cuticular regions of O. pictus demonstrates a morphological gradient. The tubercles of O. pictus resemble those of the confamilial L. vittatum , L. holtae , and H. maculosus since they exhibit micropores, but differ in that they generally are not covered by laminae. The cuticular morphology of C agilis is striking because of the numerous subdeltoid microtubercles and rugose abdominal integument. The imbricate morphology of the cephalothoracic region of C agilis is in some respects similar to that of O. pictus , although these species are not closely related. The laminae of C agilis are less prominent than those of O. pictus and are only seen in one area of its dorsum while in O. pictus they cover all regions. The diversity in the pattern of the microtubercles of C. agilis demonstrates a morphological gradient for this species. Shear (1975) described the dorsal cuticle of C. agilis as “soft and leathery without tubercles, spines, or prominent setae” in his taxonomic treatment of the Caddidae. Gruber (1974) referred to a soft cuticle with a finely granular, regularly arranged surface for C agilis. The less sclerotized cuticle of C. agilis may, with further studies, be linked to the fact that this species is restricted to very humid, densely shaded habitats. The cuticular morphology of H. kepharti is distinct because of the posteriorly oriented microtubercles and the relatively unornamented cuticular background. The setae of H. kepharti are similar to those of L. vittatum , L. holtae , H. maculosus , E. nigrum , and C. agilis but differ in that they are emitted from posterior microareolar margins at a 45 degree angle. Gruber (1970) does not specifically refer to the characteristic microtubercles of H. kepharti in his redescription of the species. Grainge and Pearson (1966) described a different cuticular morphology exhibiting numerous closely and regularly packed laminae in the related European species Nemastoma lugubre (Muller). Shear (1986) described and illustrated both the dorsal and ventral cuticular morphology of Crosbycus dasycnemus (Crosby) which he placed closest to Hesperonemastoma in MURPHREE— OPILIONID CUTICULAR MORPHOLOGY 249 his cladistic analysis of the Ceratolasmatidae. The numerous microtubercles (denticles) of the dorsal integument of C. dasycnemus resemble those of H. kepharti but were not illustrated at sufficient magnification for a detailed comparison. Shear also reported the presence of tridentate “scales” or cuticular processes on the ventral surface of C. dasycnemus which represents a marked difference in dorsal and ventral cuticular morphology. Although no descriptions of the ventral integument are given for the species in the present study, few differences in dorsal and ventral morphology were observed in specimens whose venters were examined, including that of H. kepharti. Few descriptions of the function(s) of the opilionid integument have been reported other than its physiological ability to resist desiccation (Edgar 1971) and its role in dispersing scent gland secretions discussed above. Martens (1978) indicated that the integuments of representatives of the Trogulidae, Dicranolas- matidae, and Sclerosomatinae are strikingly glandular-papillose and that soil particles adhere to the secretions of these glands producing camouflauge. It is hoped that future studies will discover still other properties and functions of the opilionid integument and its secretions. Only in recent years have taxonomic studies of opilionids included cuticular morphology as a character for analysis. Variations in integumental microstructure among families, genera, and species were used by Shear (1983, 1986) as characters for cladistic analysis. From the present study it is evident that the cuticular morphology of opilionids is a reliable character as long as adult specimens are examined, and may be used in addition to more traditional characters in systematic studies. I believe that the surface of the opilionid integument, just as Cooke and Shadab (1973) predicted for the ricinuleids, may possess a “considerable but largely untapped systematic potential.” ACKNOWLEDGMENTS I would like to thank Dr. Charles R. McGhee of Middle Tennessee State University for his guidance and encouragement during the preparation of my Master’s Thesis, a section of which is published here. I also thank Drs. Clay M. Chandler and George G. Murphy for their suggestions and support and Dr. Marion R. Wells for his assistance with scanning electron microscopy. I am grateful to Dr. William A. Shear and Thomas S. Briggs for their correspondence and assistance in identification of opilionid species and Lawrence J. Hribar for his comments on the manuscript. I also thank my wife Kay and family for their interest and encouragement throughout this investigation. LITERATURE CITED Baker, E. W. and G. W. Wharton. 1952. An Introduction to Acarology. The Macmillan Co., New York. Banks, N. 1892. A new genus of Phalangiidae. Proc. Entomol. Soc. Washington, 2:249-251. Brody, A. R. 1970. Observations on the fine structure of the developing cuticle of a soil mite Oppia coloradensis. Acarologia, 12:421-431. Byers, J. R. and C. F. Hinks. 1973. The surface sculpturing of the integument of lepidopterous larvae and its adaptive significance. Canadian J. Zool., 51:1171-1179. Cokendolpher, J. C. 1980. Replacement name for Mesosoma Weed, 1892, with a revision of the genus (Opiliones, Phalangiidae, Leiobuninae). Occas. Papers Mus., Texas Tech Univ., 66:1-19. 250 THE JOURNAL OF ARACHNOLOGY Cooke, J. A. I . and M. U. Shadab. 1973. New and little known ricinuleids of the genus Cryptoceiius (Araehnida, Ricieulei). Amer. Mus. Novitates, no. 2530. 25 pp. Crosby, C. R. and S. C. Bishop. 1924. Notes on the Opiliones of the southeastern United States with descriptions of new species. J. Elisha Mitchell Sci. Soc., 40:8-26. Crowe, J. B 1975. Studies on acarine cuticles. III. Cuticular ridges in the citrus red mite. Trans. Amer. Microscop. Soc., 94:98-108. Dalingwater, J. E. 1975. Further observations on eurypterid cuticles. Fossils and Strata, 4:271-279. Dalingwater, J. E. 1981. Studying fossil arthropod cuticles with the scanning electron microscope. Inter. Symp. Concpt. Meth. Paleo. Barcelona, pp. 319-324. Dalingwater, J. E. 1987. Chelicerate cuticle structure. Pp. 3-15, In Ecophysiology of Spiders. (W, Neetwig, ed.). Springer-Verlag, Berlin. Duffield, R. M., O. Olubajo, J. W. Wheeler and W. A. Shear. 1981. Alkylphenols in the defensive secretion of the Nearctic opilionid Stygnomma spinifera (Araehnida: Opiliones). J. Chem. Ecol, 7:445-452. Edgar, A. L. 1963. Proprioception in the legs of phalangids. Biol. Bull. Mar. Biol. Lab., Woods Hole, 124:263-267. Edgar, A. L. 1971. Studies on the biology and ecology of Michigan Phalangida (Opiliones). Misc. Pubh, Mus. ZooL, Univ. Michigan, no. 144, 64 pp. Eisner, T., A. F. Kluge, J. E. Carrel and J. Meinwald. 1971. Defense of a phalangid: liquid repellent administered by leg dabbing. Science, 1730:650-652. Eisner, T., T. H. Jones, K. Hicks, R. E. Silberglied and J. Meinwald. 1977. Quinones and phenols in the defensive secretions of neotropical opilioeids. J. Chem. Ecol, 3:321-329. Emerit, M. 1981. Sur quelques formations tegumentaires de la patte de Telema teneiia (Araign.ee, Telemidae), observees au microscope eiectronique a balayage. Atti Soc. Toscana Sci. Nat., Mem., ser. B, vol. 88, suppl, pp. 45-52. Grainge, C. A. and R. G. Pearson. 1966. Cuticular structure in the Phalangida. Nature, 211:866. Gruber, J. 1970. Die uNemastomo — ” Arten Nordamerikas (Ischyropsalidae, Opiliones, Araehnida). Ann. Naturhistor. Mus. Wien, 74:129-144. Gruber, J. 1974. Bemerkungen zur Morphologic und systematischen Stellung von Caddo, Acropsopilio und verwandter Formen (Opiliones, Araehnida). Ann. Naturhistor. Mus. Wien, 78:237-259. Hadley, N. F. 1981. Fine structure of the cuticle of the black widow spider with reference to surface lipids. Tissue and Cell, 13:805-817. Hadley, N. F. and B. K. Filshie. 1979. Fine structure of the epicuticle of the desert scorpion, Hadrurus arizonensis , with reference to the location of lipids. Tissue and Cell, 11:263-275. Hammen, L. van der. 1985. Comparative studies in Chelicerata III. Opilioeida. ZooL Verb. Leiden, no. 220:1-60. Harris, R. A. 1979. A glossary of surface sculpturing. Occas. Papers California Dept. Food Agric., Lab. Serv. in EntomoL, 28:1-31. Hill, D. E. 1979. The scales of salticid spiders. Zool. J. Line. Soc., 65:193-218. Hinton, H. E. 1970. Some little known surface structures. Pp. 41-58, In Insect Ultrastructure. (A. C. Neville, ed.). Royal Entomol. Soc., Oxford. Hoffman, R. L. 1963. A new phalangid of the genus Siw from the eastern United States, and taxonomic notes on other American siroeids. Senckenbergiana Biol., 44:129-139. Igelmund, P. 1987. Morphology, sense organs, and regeneration of the forelegs (whips) of the whip spider, Heterophrynus elaphus (Araehnida, Amblypygi). J. Morphol, 193:75-89. Immel, V. 1964. Zur Biologic und Physiologic von Nemastoma quadripunctatum (Opiliones, Dyspnoi). Zool. Jahr. Abteilung System. Oekol. Geog. Tiere., 83:129-184. Juberthie, C. and Z. Massoud. 1976. Biogeography, taxonomy and ultrastructural morphology of the cyphophthalmic Opiliones. Rev. Ecol Biol. Sol, 13:219-232. Keirans, J. E. and C. M. Clifford. 1978. The genus Ixodes in the United States: a scanning electron microscope study and key to the adults. J. Med. Entomol. Suppl., 2:1-149. Kennaugh, J. H. 1968. An examination of the cuticle of three species of Ricinulei (Araehnida). J. Zool. (London), 156:393-404. Martens, J. 1978. Spinnentiere, Araehnida. Weberknechte, Opiliones. Die Tierwelt Deutschlands, Tiel 64, Gustav Fischer Yerlag, Jena. 449 pp. Martens, J. 1979. Feinstruktur der Tarsal Druse von Siro duricorius (Joseph) (Opiliones, Sironidae). Zoomorphologie, 92:77-93. MURPHREE— OPILIONID CUTICULAR MORPHOLOGY 251 Martens, J. and W. Schawaller. 1977. Die Cheliceren-Driisen der Weberknechte nach rasteroptischen und lichtoptischen Befunden (Arachnida: Opiliones). Zoomorphologie, 86:223-250. McGhee, C. R. 1977. The politum group (bulbate species) of Leiobunum (Arachnida: Phalangida: Phalangiidae) of North America. J. Arachnol., 3:151-163. Mutvei, H. 1977. SEM studies on arthropod exoskeletons. 2. Horseshoe crab Limulus polyphemus (L.) in comparison with extinct eurypterids and recent scorpions. Zool. Scr., 6:203-213. Ope!!, B. D. 1983. The female genitalia of Hyptiotes cavatus (Araneae: Uloboridae). Trans. Amer. Microscop. Soc., 102:97-104. Pittard, K. and R. W. Mitchell. 1972. Comparative morphology of the life stages of Cryptocellus pelaezi (Arachnida, Ricinulei). Grad. Studies, Texas Tech. Univ., no. 1. 77 pp. Platnick, N. I. 1986. On the tibial and patellar glands, relationships, and American genera of the spider family Leptonetidae (Arachnida, Araneae). American Mus. Novitates, no. 2855. 16 pp. Platnick, N. I. and W. J. Gertsch. 1976. The suborders of spiders: a cladistic analysis (Arachnida, Araneae). American Mus. Novitates, no. 2607. 15 pp. Platnick, N. I. and M. U. Shadab. 1976. On Columbian Cryptocellus (Arachnida, Ricinulei). American Mus. Novitates, no. 2605. 8 pp. Quintero, D. 1975. Scanning electron microscope observations on the tarsi of the legs of amblypygids (Arachnida, Amblypygi). Proc. 6th Int. Arachnol. Congr. Amsterdam 1974:161-163. Say, T. 1821. An account of the Arachnides of the United States. J. Acad. Nat. Sci. Philadelphia, 2:59-82. Shear, W. A. 1975. The opilionid family Caddidae in North American with notes on species from other regions (Opiliones, Palpatores, Caddoidea). J. Arachnol., 2:65-88. Shear, W. A. 1980. A review of the Cyphophthalmi of the United States and Mexico, with a proposed reclassification of the suborder (Arachnida, Opiliones). American Mus. Novitates, no. 2705. 34 pp. Shear, W. A. 1986. A cladistic analysis of the Opilionid superfamily Ischyropsalidoidea, with descriptions of the new genus Acuclavella, and four new species. American Mus. Novitates, no. 2844. 29 pp. Shear, W. A. and J. Gruber. 1983. The opilionid subfamily Ortholasmatinae (Opiliones, Troguloidea, Nemastomatidae). American Mus. Novitates, no. 2757. 65 pp. Simon, E. 1879. Les Arachnides de France. VII. Contenant les orders des Chernetes, Scorpiones et Opiliones. Ann. Soc. Entomol. Beligique, 22:1-332. Spicer, G. S. 1987. Scanning electron microscopy of the palp sense organs of the harvestman Leiobunum townsendi (Arachnida: Opiliones). Trans. Am. Microsc. Soc., 106:232-239. Steinmann, H. and L. Zombori. 1981. An Atlas of Insect Morphology. Akademiai Kiado, Budapest. Torre-Bueno, J. R. 1962. A Glossary of Entomology. Brooklyn Entomol. Soc., Brooklyn. Wood, H. C. 1870. On the Phalangeae of the United States of America. Commun. Essex Inst., 6:10- 40. Woolley, T. A. 1974. The application of SEM in oribatid taxonomy. Proc. 4th Inti. Cong. Acarol., pp. 705-712. Manuscript received December 1987, revised March 1988. APPENDIX 1 Systematic list of the species included in this study. *=This species was determined to be near Erebomaster acanthina (Crosby and Bishop) but sufficiently distinct for new species status by Thomas S. Briggs of San Francisco, California. Suborder Cyphophthalmi Superfamily Sironoidea Family Sironidae Siro exilis Hoffman Suborder Laniatores Superfamily Gonyleptoidea Family Cosmetidae Vonones sayi (Simon) 252 THE JOURNAL OF ARACHNOLOGY Superfamily Travunoidea Family Cladonychiidae Erebomaster sp.* Suborder Palpatores Superfamily Phalangioidea Family Phalangiidae Subfamily Leiobuninae Leiobunum vittatum (Say) Leiobunum holtae McGhee Hadrobunus maculosus (Wood) Eumesosoma nigrum (Say) Subfamily Oligolophieae Odiellus picius (Wood) Superfamily Caddoidea Family Caddidae Caddo agilis Banks Superfamily Ischyropsalidoidea Family Ceratolasmatidae Hesperonemastoma kepharti (Crosby and Bishop) APPENDIX 2 Glossary of proposed morphological terms for describing the opilionid integument. Acicular : needle-shaped; with a long, slender point as in certain setae. Acute: sharply pointed; refers to laminar margins or other cuticular processes. Alveolus , pi. alveolae: a small depressed or cup-like cavity; refers to setal insertions or sockets. Arcuate : arched; setae that are curved like a bow. Areole, pi. areolae: a pore-like depression; refers to insertions of certain setae within rounded microtubercles. Deltoid: elongate-triangular as in certain cuticular processes; resembling the Greek letter delta with its apex extended. Dentate: toothed, with tooth-like prominences. Denticle: a tooth-like prominence; a general term. Facetodea : cuticular structures composed of numerous small facets. Foveolus , pi. foveolae: a minute pit or micropore. Glabrous: smooth; devoid of any surface features. Granulate: surfaces composed of small, obtuse to acute granules. Imbricate: cuticular laminae that partially overlap as in roof shingles or fish scales. Lamina , pi. laminae: cuticular layers, plates or scales that are generally imbricate. Micro-: precedes terms describing features that measure 0.01 mm or less in size. Mucronate: terminating in sharply pointed processes as in the margins of certain laminae. Obtuse: blunt or rounded as opposed to sharply pointed. Plicate: folded; impressed with striae to produce the appearance of having been folded or pleated. Punctate: possessing circular, concave punctures or regular depressions. Punctulate: finely punctate; with numerous small and closely set punctures or micropores. Reticulate: superficially net-like or made up of a network of elevated, angular ridges; with surface ornamentation forming polygonal areas. Rectilinear: in the form of a straight line as in certain setae. Rivulose: exhibiting small, sinuate furrows or rivulets which are not parallel. Rugose: wrinkled; refers to a pattern of impressed, irregular striae which are both parallel and intersecting producing a wrinkled appearance. Sinuate: consisting of small sinuses; refers to wavy furrows of the integument. Striae : narrow impressed lines or furrows of the integument which may be parallel or intersecting. Sub-: below; somewhat; slightly; to a lesser degree than the term it precedes. Torose: swollen; possessing superficial swellings or protuberances. Tuberculate: exhibiting rounded, projecting protuberances which may possess a micropore. 1988. The Journal of Arachnology 16:253 RESEARCH NOTES ZELOTES SANTOS (GNAPHOSIDAE, ARANEAE): DESCRIPTION OF THE MALE FROM SIERRA DE LA LAGUNA, B.C.S., MEXICO The species Zelote santos was described by Platnick and Shadab in 1983, being included in the catholicus subgroup, with only female specimens. In this paper I describe the male of Z. santos collected in the oak-pine forest at Sierra de la Laguna, B.C.S. Zelotes santos Platnick and Shadab Figs. 1-2 Two males were collected at an elevation of 1640 m in the oak-pine forest litter of Sierra de la Laguna B.C.S., Mexico, 5 March 1987 (F. Cota, A. Cota), 21 Fig. 1-2. — Zelotes santos Platnick and Shadab, male palp: 1, ventral view; 2, lateral view. 1988. The Journal of Arachnology 16:254 August 1987 (M. Vazquez). Specimens are deposited at the Arachnological Collection of the Centro de Investigaciones Biologicas de Baja California, Sur. Description. — Male: Total length 6.0-6. 3 mm; carapace 2. 6-2. 8 mm long, 2.0-2. 2 mm wide (two specimens measured). Carapace dark brown with black reticulations and bright, with long black setae, thoracic groove longitudinal; anterior eye row recurved, posterior eye row straight, diameters and interdistan- ces: AME 0.05-0.06, ALE 0.08-0.10, PME 0.06, PLE 0.08; AME-AME 0.05-0.06, AME-ALE 0.02, PME-PME 0.07, PME-PLE 0.03, ALE-PLE 0.06-0.08; MOQ length 0.20, front width 0.35, back width 0.45; chelicerae dark brown, retromargin of fang furrow with two teeth, and promargin with four; sternum with marginal brush of setae and sclerotized extensions to and between coxae. Legs dark brown with tarsi lightest, distal halves of metatarsi and tarsi scopulate; femur I 2.25 mm long with 2 dorsal macrosetae, 1 prolateral; tibia I 2.0-2. 1 mm long with 0 macrosetae; basitarsus I 1.55-1.75 mm with 2 proventral macrosetae, tibia III 1. 1-1.2 mm with one prodorsal macroseta; 3 internolaterals, 3 externolaterals, and three pairs of ventral macrosetae. Opisthosoma dark gray with shiny brown scutum, venter light yellow, spinnerets light. Palp with terminal apophysis narrow and long, fused dorsaily to embolar base, bearing a curved projection and curved embolus without prolateral hump; intercalary sclerite apparently fused with subtegulum (Figs. L2). Diagnosis. — Male: Zelotes santos seems closest to Z. union in having a low embolus, but can be distinguished by the much longer terminal apophysis. Range. — Known only from the male locality. This study has been supported by a grant from Consejo Nacional de Ciencia y Tecnologia and Secretaria de Programacion y Presupuesto, Mexico. LITERATURE CITED Platnick N. I. and M. U. Shadab. 1983. A revision of the American spiders of the genus Zelotes (Araneae: Gnaphosidae). Bull. American Mus. Nat. Hist., 1 74(2):99- 191. Maria-Luisa Jimenez, Centro de Investigaciones Biologicas de Baja California Sur, A.C., Biologia Terrestre, Apdo. Postal 128, La Paz, B.C.S., 23060, Mexico. Manuscript received September 1987, revised October 1987. TRANSITION FROM PREDATORY JUVENILE MALE TO MATE-SEARCHING ADULT IN THE ORB-WEAVING SPIDER NEPHILA CLAVIPES (ARANEAE, ARANEIDAE) Behavioral strategies of male orb-weaving spiders change rather dramatically as they mature to adulthood. Juvenile males are sedentary predators, capturing prey on webs of their own construction. However, upon reaching adulthood, they shift to a search strategy, approaching females who usually inhabit solitary webs 1988. The Journal of Arachnology 16:255 (Robinson 1982). As pronounced as these changes are, few field data have been gathered on marked, unrestrained juvenile males as they mature to adulthood. The purpose of this study is to provide ethological descriptions of this transition phase in the life cycle of Nephila clavipes , a New World orb-weaver. Specifically, we describe web maintenance, changes in body coloration and size, sperm web construction, sperm transfer to the palps, and dispersal as males mature from the penultimate instar to adulthood. Forty-one juvenile males in the penultimate instar were observed from the second week in July to the last week in August, 1984 at the F. Edward Herbert Center of Tulane University, located about 20 km south of New Orleans, Louisiana. Criteria for inclusion were that the male must have been residing on a male-constructed web, and there had to be evidence of the final molt during the course of observation. These criteria were met by 28 males. A census of the males was taken every morning between 0800 and 1000 h. For individual identification, each was marked on the dorsum of the abdomen with fast-drying acrylic paint. Data collected include: estimated amount of viscid spiral in good repair, number of sperm webs, abdominal coloration, and occurrence of molt. The day of the molt was determined by the absence of a paint mark on a recently molted male inhabiting an identified web and/or the presence of a paint- marked exoskeleton. Subjects were re-marked after molting and checked daily until they abandoned the web. The adjacent forest area was searched daily for marked males. To quantify the change in size occurring at the final molt, we examined eight unrestrained, unmarked males found outside of the census area who had just molted to adulthood. The exoskeleton as well as a front leg (I) were removed and the tibia-patella length of the leg and corresponding portion of the exoskeleton was measured (following Vollrath 1983). For comparison, legs and exoskeletons of 17 females maturing to adulthood were measured in a similar manner. By the day of the final molt, males had allowed their viscid spirals to almost totally deteriorate (Table 1), as do females at the final molt (Christenson et al. 1985). However, males did not construct a stabilimentum on the final web, as do females (Robinson and Robinson 1973). The length of the tibia-patella portion of the male front leg increased by 21.5% (SD = 8.8) from the penultimate to the final instar. This was significantly less than the rate of growth for maturing females (35.8%, SD 10.2; F = 11.445, df 1,23, p - 0.003). Males built their first sperm web an average of 2.1 days (SD = 0.55) after molting. The typical sperm web was trapezoidal, about 5X5 mm in size, and located in the barrier strands or remains of the viscid spiral. Two males were observed constructing sperm webs. First, they moved the abdomen back and forth between what appeared to be already established silk strands, for 150 s in one case and 270 s in the other. To the unaided eye, the resulting web appeared as a dense mat of fine strands. The genital opening was then moved against the web with one male bouncing and the other pushing the ventral abdomen onto the web. Sperm deposition took 75 and 60 seconds. Very quickly thereafter the males began dipping the palps onto the web with the conductor held parallel to the web plane. They were dipped in a mostly alternating manner, once every two seconds, for 105 and 135 seconds. Microscopic examination of ten sperm webs revealed 1988. The Journal of Arachnology 16:256 Table 1. — Census data relating to condition of the viscid spiral and presence of sperm webs on the orbs of male N. clavipes gathered during a ten day period around the time of their final molt. Day N % of Viscid Spiral Intact >90% <10% % of Orbs with Sperm Webs Of Orbs with Sperm Webs, X Number Present Range of Sperm Webs PRE-MOLT 5 17 0.94 0.06 0 — 4 19 0.89 0.11 0 — 3 21 0.52 0.34 0 — — 2 25 0.40 0.36 0 — 1 25 0.12 0.79 0 — — Molt 28 0.04 0.93 0 Hfiili — POST-MOLT 1 25 0.00 1.0 12.0 1.0 i-i 2 19 0.00 1.0 68.4 2.2 1-5 3 8 0.00 1.0 75.0 1.5 1-2 4 3 0.00 1.0 100.0 2.0 1-3 the strands to be loose and tangled in appearance. Transfer to the palps must be quite efficient because only one web contained sperm, and it had only one sperm. Frequently, we found several sperm webs on a given orb (Table 1). Unfortunately, it was not possible to accurately determine the total number of sperm webs constructed by a given male. They were damaged by wind and rain and thus nearly impossible to individually recognize from day to day. During this time male color is changing. Typical abdomen coloration of the juvenile male was female-like, yellow/ caramel and white (see Levi 1980). In the penultimate instar, palps and femurs were a translucent light grey. However, on the day after the final molt, abdomen coloration was darker for half of the subjects, palps darker in all males, and femurs darker in 86%. By the third day, 20% of the males had the typical adult abdominal coloration, a uniform dark caramel or copper. The black midline stripe and the black elongated marks lateral to posterior abdomen, prominent in juveniles, were visible but relatively faint. Adult coloration was complete in about one week. These color changes were more pronounced than those of the female, for they maintain the yellow juvenile coloration. Robinson and Robinson (1976) noted that the functions of maturational color changes are not clear. It is possible that the relatively dark color of the male serves as camouflage while moving on branches or twigs when between female webs. Predatory pressure on moving males is thought to be relatively high (Christenson & Goist 1979; Vollrath 1980). Whatever their significances, changes in coloration are indispensible in estimating male age. Males abandoned the web an average of three days (SD = 0.74) after molting and an average of one day (SD = 0.74) after appearance of the first sperm web. Eight of our 28 males were found on female webs, six with a juvenile female and two with a female who had just molted. It should be noted that most females in the immediate area were juveniles. Only one male had moved to the nearest female; their webs were already connected. Although mating occurs on the female’s web, sperm webs are infrequently found there. In July, out of a total of 770 census-days of marked males on female 1988. The Journal of Arachnology 16:257 webs, only seven sperm webs were noted. Census data gathered in a similar manner in 1982 indicated 17 sperm webs on female orbs in a total of 628 marked male-days. In three of these cases, the male present did not yet have the adult coloration and probably had not had the opportunity to build sperm webs on its own orb prior to abandonment. The failure of sperm webs to appear after mating is consistent with the observation that male N. clavipes deplete their sperms stores after mating with a female just after her final molt and do not produce more sperm (Manuscript in preparation). We thank J. Coddington, A. Rypstra, and B. Robinson for their comments. This study was funded by NSF grant BNS-83 17988 to the second author. LITERATURE CITED Christenson, T., S. G. Brown, P. A. Wenzl, E. M. Hill and K. C. Goist. 1985. Mating behavior of the golden-orb-weavieg spider, Nephila clavipes: I. Female receptivity and male courtship. J. Comp. Psych., 99:160-166. Christenson, T. and K. C. Goist. 1979. Costs and benefits of male-male competition in the orb weaving spider, Nephila clavipes. Behav. Ecol. Sociobiol., 5:87-92. Levi, H. W. 1980. The orb-weaver genus Mecynogea, the Subfamily Metinae and the genera Pachygnatha, Glenognatha and Azilia of the Subfamily Tetragnathinae North of Mexico (Araneae: Araneidae). Bull. Mus. Comp. Zool., 149:1-75. Robinson, M. H. 1982. Courtship and mating behavior in spiders. Ann. Rev. Entomol., 27:1-20. Robinson, M. H. and B. C. Robinson. 1973. Ecology and behavior of the giant wood spider Nephila maculata (Fabricius) in New Guinea. Smithson. Contrib. Zool., 149:1-76. Robinson, M. H. and B. C. Robinson. 1974. Adaptive complexity: The thermoregulatory postures of the golden-web spider, Nephila clavipes , at low latitudes. American Midi. Nat., 92:386-396. Robinson, M. H. and B. Robinson. 1976. The ecology and behavior of Nephila maculata : A supplement. Smithson. Contrib. Zool., 218:1-22. Vollrath, F. 1980. Male body size and fitness in the web-building spider, Nephila clavipes. Z. Tierpsychol., 53:61-78. Vollrath, F. 1983. Relative and absolute growth in Nephila clavipes (Arachnida: Araneae: Argiopidae). Verh. Naturwiss. Ver. Hamburg, 26:277-289. Leann Myers and Terry Christenson, Department of Psychology, Tulane University, New Orleans, Louisiana 70118 USA. Manuscript received April 1987, revised November 1987. MALE RESIDENCY ON JUVENILE FEMALE NEPHILA CLAVIPES (ARANEAE, ARANEIDAE) WEBS Male orb-weaving Nephila clavipes leave their own individually-constructed orbs after the final molt and move about in search of mates. Males are likely to land on webs of females of various instars for it appears that they are not attracted to webs of sexually receptive females by distance-acting pheromones (Christenson et al. 1985). Once on the female’s web, duration of male residency might be related to female instar because adult males are often found on webs of 1988. The Journal of Arachnology 16:258 larger than smaller juvenile females (Farr 1976; Brown et al. 1985). First, we asked if marked, unrestrained male N. clavipes remain longer on a web inhabited by a female in the penultimate instar, approaching sexual receptivity, than on a web of a relatively smaller juvenile, three or four instars from adulthood. We found that they do remain longer with the female in the penultimate instar. Female responsiveness could be one factor underlying this variation in male residency. If so, one would expect juvenile females of different instars to respond differently to adult males. Second, we placed males onto juvenile female webs to determine if females in the penultimate instar respond less aggressively than females in earlier instars. Observations were conducted at the F. Edward Herbert Center of Tulane University, located about 20 km south of New Orleans, Louisiana. To determine if duration of male residency on the web is affected by female instar, we examined census data gathered during July and August, 1980 and 1982 on paint- marked males, small juvenile females (12-15.5 mm in cephalothorax-abdomen length), and females in the penultimate instar (18-23 mm). Females in the latter group were observed to molt once and mate. To ensure that the duration of male residency was most likely determined by the subject male, we analyzed only those cases ( N - 36) in which the female remained on the web after the male had departed and in which no other male had come onto the web, possibly displacing the male in question. Female response to an added male was observed during the mating season, from July to mid-August, 1980. Juvenile males in the penultimate instar were placed in Fiberglas-screened enclosures situated in the field and fed Drosophila daily. No sooner than four days after their final molt, twenty males were paint- marked and randomly assigned to a relatively small juvenile female (12-14 mm) or a female in the penultimate instar (18-20 mm). Other males already present on the females’ webs were removed and the subject males were transported on a thin stick and gently placed on barrier strands, about 30 cm from the hub. Male and female behaviors listed in Table 1 were recorded serially for 20 minutes. Three males placed with small juveniles moved onto adjacent foliage within the first minute. In two cases this occurred before the female had made any behavioral response. In the other case the female response, one pluck, did not immediately precede the male’s departure. A fourth male spent most of the observation period under a leaf at a silk attachment point. Data for these four males were excluded from statistical analyses. Analysis of census data revealed that marked, unrestrained males remained an average of 2.6 days (SD = 2.6) with small juvenile females whereas males remained 9.1 days (SD = 5.2) with females in the penultimate instar i f = 17.84, df = 1,34, p - 0.0001). After the male was added to the web, large and small juvenile females strand- plucked while at the hub with about equal frequency (Table 1). However, the small females oriented to the male, strand-plucked while oriented, and chased the male more frequently (Table 1). Males slowly approached while probing both sizes of females with about equal frequency. Males with larger females spent more time within 10 cm of the female (x = 620.2 s versus x = 242.5 s; F = 5.343, df 1,14, p - 0.002), abdomen vibrated and probed while stationary more frequently (Table 1), and, by the end of the 20 minute observation period, were more likely to be within 10 cm of the hub (9/10 LG, 3/7 SM; Chi Square = 4.41, p = 0.036) than the males with smaller females. Most of the males of both groups remained for at 1988. The Journal of Arachnology 16:259 Table 1. — Behavior of small juvenile N. clavipes females (12-14 mm) and larger juvenile females in the penultimate instar (18-20 mm) and the adult males placed on webs of these females during a 20 minute serial record, a = multiple occurrences without a return to hub position were scored as one event; b = each pluck scored as a separate event; c = often occuring repeatedly in a prolonged sequence which was scored as one event; intermittent sequences separated by 5 s were scored as multiple events; d = bouts of rapid vibration of the posterior tip of the abdomen; and e = a sweeping, waving of the l’s while stationary in barrier strands. Small Juvenile {n- 6) Large Juvenile (n- 10) X SD X SD F P FEMALE BEHAVIOR Orient to male3 8.5 5.8 4.0 2.9 4.261 0.058 Strand-pluck in hub position15 4.7 2.3 5.5 5.5 0.120 0.734 Pluck while oriented to maleb 14.2 12.1 5.0 4.4 4.854 0.045 Approach male3 3.3 4.1 0.9 1.4 3.101 0.100 Chase male3 1.3 1.4 0.2 0.6 5.214 0.039 MALE BEHAVIOR Slow approach and probec 38.3 22.9 38.5 19.1 0.001 0.988 Probe with l’s while stationary1" 9.7 9.7 63.6 52.9 5.943 0.029 Abdominal vibration6’61 0.3 0.8 16.2 14.4 7.118 0.018 Lateral leg sweeps0’6 2.7 4.8 1.9 1.8 0.211 0.653 least 24 hours (6/9 S, 8/9 L). This does not necessarily exclude female responsiveness as a factor contributing to duration of residency. Census data show that males on the webs of smaller females remained, on the average, two and one half days. Duration of male residency beyond the first day was difficult to assess for the added males because some females abandoned the web before the male did or a larger male came onto the web, possibly displacing the subject male. We conclude that unrestrained males spend relatively more time on the web of a female approaching her final molt, which is probably adaptive for the male. We suggest that female response to the male might be one factor underlying this variation in male residency. We thank R. Suter and an anonymous reviewer for comments. LITERATURE CITED Brown, S., E. M. Hill, K. C. Goist, P. A. Wenzl and T. Christenson. 1985. Ecological and seasonal variation in a free-moving population of the golden-web spider, Nephila clavipes. Bull. British Arachnol. Soc., 6:313-319. Christenson, T., S. G. Brown, P. A. Wenzl, E. M. Hill and K. C. Goist. 1985. Mating behavior of the golden orb-weaving spider, Nephila clavipes'. I. Female receptivity and male courtship. J. Comp. Psych., 99:160-166. Farr, J. A. 1976. Social behavior of the golden silk spider, Nephila clavipes. J. Arachnol., 4:137-144. Elizabeth M. Hill, Department of Psychology, Tulane University, New Orleans, Louisiana 70118 USA (Present address: Evolution and Human Behavior Program, 1524 Rackham Bldg., University of Michigan, Ann Arbor, Michigan 48109 USA); and Terry Christenson, Department of Psychology, Tulane University, New Orleans, Louisiana 70118 USA. Manuscript received April 1987, revised November 1987. 1988. The Journal of Arachnology 16:260 NATURAL HISTORY OBSERVATIONS OF SALTICUS A USTINENSIS (ARANEAE, SALTICIDAE) IN NORTH-CENTRAL TEXAS The zebra spider, Salticus austinensis Gertsch, is a small jumping spider reported only from Texas. In north-central Texas it is widely distributed and in suitable habitats may reach high population densities. The localized density of this zebra spider as well as its dirunal habits, preference for open, exposed foraging grounds, and conspicuously bold black-and-white markings, makes it an ideal subject for field behavioral studies. Although some behavioral investigations into the closely-related S. scenicus (Clerck) are available (Jacques and Dill 1980, Am. Nat. 116:899-901), no such information exists for S. austinensis. A population of Salticus austinensis inhabiting the outside walls of a brick veneer home in Wichita Fails, Texas, constituted our original sample. Observations of the spiders’ activities from April through July 1986 were recorded on an almost daily basis for periods varying from ten minutes to four hours. A captive population of 18 adult zebra spiders (2 males, 16 females) was maintained in a single cage measuring 40 X 20 X 25 cm for further observations and collection of reproductive data. Field observations were conducted in Archer, Baylor, Clay, and Wichita counties of north-central Texas. The following generalizations regarding aspects of the natural history of S. austinensis are based on our observations under both natural and controlled situations. Periods of activity. — Zebra spiders exhibited more restricted periods of active foraging than other species of sympatric hunting spiders (i.e., Phidippus audax , Metacryha taenolia , Platycryptus undatus and Metaphidippus imm.) and were usually the last to appear and first to retire. Although we have sightings from shortly after sunrise to one hour before sundown, most sightings were during the hours of maximum light from ca 1000 h to 1500 h. Since Salticus , like many species of spiders, is not a daily forager, the maximum number of sighted adult individuals at our original site fluctuated considerably. Temperatures during our period of observations ranged from near 22° to 40° C, and seemed to have no major effect on spider activity. However, overcast and rainy days resulted in noticeably lessened activity. Winds seemed to curtail foraging on some days, especially when gusts exceeded 32 kmp. Distribution and habitat. — Carpenter (1972, Southwestern Nat. 17(2): 161-168) noted that the zebra spider in Wichita County was restricted largely to vertical surfaces such as tree trunks and walls of buildings. Throughout our study area, we found this to be true, although overhanging surfaces of buildings and rock cliffs were equally suitable. Preferred foraging surfaces were relatively smooth, exposed, and well-lit, presumably to avoid ambush by predators. Single spiders were occasionally noted foraging along tree trunks or collected by beating shrubs and trees. Salticus austinensis may well be widely distributed in sparse numbers at such sites. Greatly increased densities of S. austinensis (surpassing that of other spider species combined) was predictable in our study area where large areas of open foraging surfaces were available (e.g., walls of buildings, rock-faced cliffs, concrete dams), and wherever their preferred species of prey (midges) were abundant. Midges are aquatic breeders and occur in large populations near water. Foraging surfaces along shorelines of lakes, streams, and 1988. The Journal of Arachnology 16:261 large stock ponds, or surfaces farther from the water (but still within flight range of midges) that are illuminated at night and consequently attract midges, permit sizeable Salticus populations. Prey species. — Forty-six prey items were randomly removed from feeding spiders. The majority of the prey were chironomid midges (74%). These were followed by mosquitoes (11%), small lepidoptera (9%), two small dipterans (4%) and a small beetle. The largest observed prey was a house fly. The chironomids comprised the majority of prey due to their abundance and their ease of capture. Population biology. — Sex ratios heavily favored females. Males, readily characterized by their more slender build and elongate, dark chelicerae, never exceeded 10 percent of any observed population. Because of this sex bias, and because zebra spiders do not usually emerge daily, males often were not detected at observation sites. Intraspecific interactions are characterized by mutual avoidance, although territorial displays between adult males were observed on two occasions. In the first such instance, the spiders met head-to-head with chelicerae and pedipalps oriented laterally at nearly a 180-degree angle for about five seconds before mutual retreat. The second encounter occurred within a small collecting vial in which two specimens were held. This interaction was only observed during the latter stage. One male assumed an immobile (and presumably submissive) posture while the second animal with fully extended chelicerae and pedipalps approached from the left side. Contact was maintained for several seconds before retreat by the aggressor. The two spiders remained indifferent to each other and were subsequently transferred into the population cage. Jacques and Dill (1980, Am. Nat. 116:899-901) record intraspecific encounters between Salticus scenicus but do not specify the sexes of their specimens. We did not observe the hibernacula (webbed shelters) of Salticus austinensis under natural conditions, as they are apparently in available cracks and crevices in and around foraging grounds. A small crack in the overhang of our original study site was the entrance to overnight shelter for several adult spiders, which were noted to emerge from it, often within seconds of each other. Hibernacula of captive individuals appeared to be randomly dispersed in the population cage. Although there is no evidence that communal denning commonly occurs, we speculate that such may be the case in instances where suitable shelter for hibernacula is scarce. However, this would appear to be more a case of opportunistic behavior than of true sociality. Reproductive potential of Salticus austinensis is low. Of seven egg clutches laid by captive specimens, the range of egg/ clutch is two to five (mean, 3.6). Interspecific relationships. — Zebra spiders carefully avoid all contact with other species of spiders, regardless of size. During foraging, zebra spiders carefully skirted webs of various sizes and species of theridiids. Observed predation by other spiders on S. austinensis was a rare event. A large Phidippus audax (Hentz) was observed feeding on an adult female zebra spider. On two occasions, adult female Salticus were observed in the webs of theridiids, one was dead and the other was still attempting to escape. These webs appeared abandoned, as they were vacated and cluttered with debris, and the trapping was apparently accidental. Several mud-daubers, both Sceliphron caementarium (Drury) and Chalybion californicum (Saussure), were noted near some of our study sites, but 1988. The Journal of Arachnology 16:262 examination of the nests revealed oxyopids, thomisids, and a single Platycryptus to be the prey of these wasps. At sites where S. austinensis was the most abundant species, the second most commonly found spider was the larger salticid, Platycryptus undatus (De Geer). The two species appear to occupy similar niches, although P. undatus often reside in exposed hibernacula and appear to be less active foragers. The two species exhibit mutual avoidance. On several occasions, the larger Platycryptus was attracted by the movement of a foraging zebra spider, but would never approach. We once confined adult females of each species together in a small plastic vial for 24 hours in an attempt to induce agonistic behavior, but none was observed. In summary, some aspects of the population biology of the zebra spider, Salticus austinensis , appear unusual for the family Salticidae, and deserve further study: — ecological and behavioral relationships among Salticus individuals and between Salticus and Platycryptus undatus ; — indicated low reproductive potential of this species, so conspicuous in markings and foraging behavior, and therefore presumably more prone to predation. Norman V. Horner, Frederick B. Stangl, Jr., and G. Kip Fuller, Department of Biology, Midwestern State University, Wichita Falls, Texas 76308 USA. Manuscript received October 1987, revised December 1987. FUNCTIONAL ASPECTS OF THE MALE PALPAL ORGAN IN DOLOMEDES TENEBROSUS , WITH NOTES ON THE MATING BEHAVIOR (ARANEAE, PISAURIDAE) In this note, we describe a locking mechanism in the male palp of Dolomedes tenebrosus Hentz, 1843, and include notes on mating behavior of the species. In numerous spider families, the adult males possess a conspicuous tibial apophysis. These tibial apophyses occur in a great diversity of shape and form. They are often species-typical and frequently figured in taxonomic works to facilitate identification. Their function, however, is unknown. While observing the copulation of a pair of D. tenebrosus , we were able to preserve the male palp in the naturally expanded stage. The investigation of the palp provided an insight into the functional complex of the expanded genital bulb and tibial apophysis. Thus far, the European Dolomedes fimbriatus (Clerck, 1758) is the only species of this large genus for which mating behavior and copulatory position are well known (Bonnet 1924; Gerhardt 1926). The male of D. fimbriatus displays a courtship consisting of rapidly waving his front legs and extending the pedipalps. The female postures in a specific position: all legs are held close to the body, and the patellae touch each other above the prosoma (Schmidt 1957). The male mates with the female on the ground or in the vegetation, using copulatory position II 1988. The Journal of Arachnology 16:263 (Le.. male on female’s dorsum, facing in opposite direction, his prosoma over her abdomen), typical of the “modern hunting spiders” (Gerhardt 1924). Only one copulation of D. tenebrosus was observed in the laboratory. The female (27 mm body length; collected Lynchburg, Virginia) had molted 14 days before, and the male (9 mm body length; collected Washington, D.C.) 72 days before the observation was made. The male was stimulated with silk threads made by the female, which were placed in his cage four hours prior to copulation. After dark the pair was placed on an arch formed of 25 mm wire mesh. The behavior was recorded on videotape with a Panasonic WV1854 video camera, using infrared light (> 800 nm wavelength). The female terminated the copulation by killing the male. We retrieved his body (with the right palp still expanded) and preserved it in 80% ethanol. Courtship and copulation lasted 1.5 hours. The female was placed onto the wire screen, where she moved around for a few minutes and finally assumed a “ventral-up” position. The male was placed onto the wire screen at a distance of approximately 15 cm from the female. He waved with the outstretched front legs, contacted her silk lines, and approached the female. After the male had initially contacted the female and stroked her I and II legs, the female groomed these legs vigorously. During 50 minutes of courtship, the male lightly stroked and tapped the distal segments of the female’s anterior legs and proceeded to stroke her abdomen. The female remained mostly motionless when contacted by the male. The male climbed on the female’s dorsum, their bodies parallel but pointing in opposite directions, as if anticipating copulatory position II. The female pulled her legs closer to the body; legs III and IV were not in contact with the wire screen. The male approached the female’s venter from both sides, about 35 times in total. The female responded to each attempt by rocking her venter laterally toward the male, thus providing more room for the male to approach her epigynal area. This phase lasted for 32 minutes; tempo and frequency of the male’s attempts increased during that time. Copulation itself lasted about 4.5 minutes. The male abruptly passed completely across her right side and onto her venter in a perpendicular position, inserted his right palp into her right copulatory pore (Fig. 1), and simultaneously expanded the basal and median hematodochae. During the insertion of the palp, we observed no hematodochal pulsing. Both animals were still. The female slowly pulled the male’s body with her front legs into a parallel orientation to hers, juxtaposing his abdomen to her mouth. When the female bit the tip of the male’s abdomen the palp sprang free of the epigynum almost immediately, and remained, in an expanded state, still attached to the male’s body. At this point we retrieved the male’s body. A study of the expanded bulb revealed that a heavily sclerotized part of the embolic division fitted behind the tibial apophysis, and apparently arrested the rotation of the bulb. Figure 2 shows the expanded right palp in retrolateral view. The sclerites of the genital bulb are labelled according to Comstock’s nomenclature (1910:180) used for Dolomedes scriptus Hentz, 1845. Attached to the distal end of the tegulum by an inflatable membrane is a strongly sclerotized tube. At its distal tip, this sclerotized tube bears the fulcrum, the lateral subterminal apophysis and the spiral embolus. During expansion, the membrane connecting the tegulum and the sclerotized tube is inflated and the sclerotized tube assumes an erect position. Due to the inflation 1988. The Journal of Aracheology 16:264 Fig. 1. — Copulatory position of D. tenebrosus , drawn from videotape. Male drawn in black. Scale = I cm. and rotation of the basal and median hematodoehae, the subtegulum-tegulum- complex is tilted towards the retrolateral side of the palp. In this position, the sclerotized tube fits snugly behind the tibial apophysis and arrests the rotation of the bulb. The described locking mechanism proved to be strong, and even repeated handling of the palp did not release the genital bulb from its arrested position. Fig. 2. — Dorsal view of expanded palp of D. tenebrosus. Bulb is ‘locked” behind tibial apophysis. 1. st. a. = lateral subterminal apophysis. Scale = 1 mm. 1988. The Journal of Arachnology 16:265 The observation of an internal locking mechanism in the male palp during expansion sheds new light on the function of male tibial apophyses. Most genera currently assigned to the Pisauridae (or Pisauridae and Dolomedidae; Lehtinen 1967) possess well-developed and often large tibial apophyses. In many cases, they provide useful species-specific characters. Heimer (1982) described internal locking mechanisms in Theridiidae and Linyphiidae, in which the paracymbium and different parts of the bulb form a functional complex during copulation that arrests the rotation of the bulb. The locking mechanism in D. tenehrosus seems functionally similar although the structural elements of the mechanisms are not homologous. The copulatory position of D. tenebrosus appears to be modified from the standard copulatory position II in D. fimbriatus , where the males are more similar in size to the females. The stroking motion of the male resembled the leg waving motion observed in D. fimbriatus , D, scriptus, D. vittatus Walckenaer, 1837, and D. triton (Walckenaer, 1837) (see Carico 1973; Roland & Rovner 1983). The female D. tenebrosus pulled her legs close to her body as if she were about to assume a posture similar to females of D. fimbriatus. D. scriptus and D. vittatus do not pull their legs close to the body while mating and the mating position is modified as well (Carico 1973). This study was conducted during a postdoctoral fellowship of the first author at the National Museum of Natural History, Smithsonian Institution in Washington. We thank Dr. J. E. Carico and Dr. C. D. Dondale for critically reading the manuscript. LITERATURE CITED Bonnet, P. 1924. Sur l’accouplement de Dolomedes fimbriatus Cl. (Araneides). C. R. Soc. biol., 91:437-438. Carico, J. E. 1973. The nearctic species of the genus Dolomedes (Araneae: Pisauridae). Bull. Mus. Comp. Zool, 144:435-488. Comstock, J. H. 1910. The palpi of male spiders. Ann. Entomol. Soc. Amer., 3:161-185. Gerhardt, U. 1924. Weitere Studien fiber die Biologic der Spinnen. Arch. f. Naturgesch., 90, Abt. A:85-192. Gerhardt, U. 1926. Weitere Untersuchungen zur Biologie der Spinnen. Z. Morph. Okol. Tiere, 6:1-77. Heimer, S. 1982. Interne Arretierungsmechanismen an den Kopulationsorganen mannlicher Spinnen (Arachnida, Araneae). Entomol. Abh., 45:35-64. Lehtinen, P. T. 1967. Classification of the cribellate spiders and some allied families. Ann. Zool. Fennici, 4:199-468. Roland, C. and J. S. Rovner. 1983. Chemical and vibratory communication in the aquatic pisaurid spider Dolomedes triton (Araneae: Pisauridae). J. Arachnol., 11:77-85. Schmidt, G. 1957. Einige Notizen fiber Dolomedes fimbriatus (CL.). Zool. Anz., 158:888-897. Petra Sierwald and Jonathan A. Coddington, Department of Entomology, National Museum of Natural History, Smithsonian Institution, Washington, DC 20560 USA. Manuscript received December 1987, revised February 1988. 1988. The Journal of Arachnology 16:266 ANANTERIS FESTAE BORELLI, ESPECE DE SCORPION CARACTERXSTIQUE DU CENTRE D’ENDEMISME “CHIMBORAZO” EN EQUATEUR Des sa creation par Thorell, en 1891, le genre Ananteris s’est caracterise par un nombre reduit d’especes considerees comme tres rares dans leur ensemble. II est vrai que 62 ans se sont ecoules entre la description de la troisieme espece, Ananteris cussinii Borelli, 1910 et celle de la quatrieme espece, Ananteris venezuelensis Gonzalez-Sponga, 1972. Le travail de revision globale du genre (Lourengo, W. R., 1982, Bull. Mus. natn. Hist, nat., Paris, 4e ser., 4:119-151) apporte finalement une contribution plus large a la taxonomie et a la repartition geographique du groupe. Aujourd’hui le nombre d’especes connues d 'Ananteris (15) s’est considerable- ment accru et me me si plusieurs d’entre elles demeurent peu connues, d’autres constituent des cas d’endemisme assez nets. Un exemple tres interessant de rarete d’une espece est celui d Ananteris festae, decrite par Borelli en 1899 sur un seul exemplaire femelle du Rio Peripa. Jusqu’a la revision du genre, cet exemplaire etait le seul connu pour l’espece. En 1982 deux autres exemplaires sont cites par Lourengo, un male et une femelle collectes a Rio Palenque, a 50 Km de Quevedo. Fig. 1. — Repartition d "Ananteris festae en Equateur, en correlation avec le centre d’endemisme “CHIMBORAZO”. 1988. The Journal of Arachnology 16:267 La rarete d’une espece comme A. festae qui habile un milieu forestier dans la litiere est principalement due a deux facteurs: sa condition d’animal cryptique et sa petite taille; les males ne depassent pas 15 mm et les femelles 20 mm. Ainsi toute chasse a vue n’est pas rentable. A present 1’etude d’un plus grand nombre d’exemplaires collectes par des methodes d ’extraction de Berlese permet de mieux connaitre la repartition de cette espece endemique pour le centre-ouest de l’Equateur (Fig. 1). Ananteris festae presente une distribution endemique qui correspond tres bien au centre d’endemisme “CHIMBORAZO” defini d’apres 1’etude des Papillons Heliconiini (Brown, K. S., 1979, Tese, Univ. Est. Campinas, Bresil, 265 p). Materiel examine. — EQUATEUR: LOS RIOS: Rio Palenque, 50 Km de Quevedo, 1 male, 1 femelle (RA), R. Alsina coll. CCRP, 1 janvier 1981, 2 males, 1 femelle (JB), S. Sandoval coll. Mars 1981, 1 femelle (JB), S. Sandoval coll, 26 decembre 1980, 1 male, 1 femelle (JB), S. Sandoval coll. PECHINCHA: Rio Peripa, 1895-98, 1 femelle-holotype (MIZSUT-Sc-5-274), L. Festa coll. 4 Km Sto. Domingo, 8 juin 1976, 1 male (FMNH), S. Peck coll. (Ber.-342, termite nest), 16 Km SE Sto. Domingo, 15 juin 1975, 1 femelle (FMNH), S. Peck coll. (Ber.-300, leaf litter), 47 Km S Sto. Domingo, Rio Palenque, 18 mai 1975, 2 males, 3 femelles (FMNH), S. Peck coll. (Ber. B-299A, forest litter), 25 fevrier 1976, 1 femelle (FMNH), S. Peck coll. (Ber., decaying fruit). Wilson R. Louren^o, Laboratoire de Zoologie (Arthropodes), Museum National d’Histoire Naturelle, 61, rue de Buffon, F-75231 Paris Cedex 05, France. Manuscript received , accepted January 1988. A COMMON METHOD OF SOUND PRODUCTION BY COURTING JUMPING SPIDERS (ARANEAE, SALTICIDAE) There have been only a few reports of sound production by salticids (Bristowe 1929; Edwards 1981; Maddison 1982; Gwynne and Dadour 1985), spiders which have been thought to rely heavily on visual communication (Jackson 1982:246). Our recent recordings of jumping spider courtship have now confirmed that the behavior of abdomen twitching, widespread in the family, produces a sound, as anticipated by Jackson (1978, 1982:218), which is easily recorded and possibly significant. To record both sound and behavior, spiders were placed on a piece of light cardboard taped over a Pressure Zone Microphone® (“Sound Grabber”, Crown, Inc.) connected to a Pentax™ Video Recorder which also received video input from a JVC™ color video camera (Model 6X-N74™ with 105 mm macro lens). Eighteen North American species were recorded: six Habronattus species (see Maddison and Stratton 1988), Maevia inclemens (Walckenaer), the dendryphantines Eris aurantia (Lucas), Erls limbata (Banks), Metaphidippus watonus Chamberlin & Ivie, M. cf. manni (Peckham & Peckham), M. cf. galathea (Walckenaer), Phidippus cf. comatus Peckham & Peckham, Sassacus papenhoei Peckham & Peckham, Tutelina elegans (Hentz), and T. formicaria (Emerton), and the euophryines Habrocestum pulex (Hentz) and Tylogonus 1988. The Journal of Arachnology 16:268 Figs. 1-3. — Sonagrams of sounds made by abdominal twitching during dendryphantine courtship: L Metaphidippus cf. manni , showing sounds from three abdominal twitches; 2, Phidippus cf. comatus (the vertical streaks at left result from the palp hitting the substrate; the dark spot at right results from abdomen twitching); 3, Sassacus papenhoei , showing sounds from 1 1 abdominal twitches. Analyzed using a Kay Sonagraph 606 IB®. morosus (Peckham & Peckham). These species are ail found frequently on foliage or leaf litter. In nine of these species, E. aurantia , M. wutonus , M. cf. manni , P. cf. comatus , S. papenhoei , Hahronattus cognatus, H. conjunctus , H. elegans , H. borealis , the males would occasionally twitch the abdomen down and up during courtship, at the same time emitting a buzzing or purring sound at frequencies mostly below 500 Hz (Figures 1-3; suitable sonagrams were not obtained for M. watonus and E. aurantia). Though one would have expected the abdominal twitches to generate some vibrations, it was surprising that they were strong enough to be recorded as airborne sounds by our relatively crude equipment. The other species were not seen to twitch the abdomen nor were they heard to make such noises, except one subadult female of Eris limbata who buzzed her abdomen while a male was courting. In all species the abdomen contacts neither the substrate nor the carapace while twitching. The sound may be produced by the legs recoiling and striking the substratum on each of the abdominal twitches, although in most species these twitches appear gentle. Because this abdominal twitching is hidden and seems unlikely to function as a visual stimulus to the female (Jackson 1982), if it has a communicatory function at all it is probably via the vibrations produced and transmitted through the substrate, though this has yet to be tested experimentally. Given the ubiquity of abdominal twitching in salticid courtship, it therefore appears that acoustic communication in salticids may be the rule, rather than the exception. 1988. The Journal of Arachnology 16:269 We would like to thank Arthur Schwartz for use of sound treated room, R. J. O’Hara for assistance with the Sonograph, and H. W. Levi and David Maddison for commenting on the paper. Specimens are deposited in the Museum of Comparative Zoology. LITERATURE CITED Bristowe, W. S. 1929. The mating habits of spiders, with special reference to the problems surrounding sex dimorphism. Proc. Zool. Soc. London, (2) 1929:309-358. Edwards, G. B. 1981. Sound production by courting males of Phidippus mystaceus (Araneae: Salticidae). Psyche, 88:199-214. Gwynne, D. T. and I. R. Dadour. 1985. A new mechanism of sound production by courting male jumping spiders (Araneae: Salticidae, Saitis michaelseni Simon). J. Zoology, London (A), 207:35- 42. Jackson, R. R. 1978. An analysis of alternative mating tactics of the jumping spider Phidippus johnsoni (Araneae, Salticidae). J. Arachnol., 5:185-230. Jackson, R. R. 1982. The behavior of communicating in jumping spiders (Salticidae). Pp. 213-247, In Spider Communication: Mechanisms and Ecological Significance. (P. N. Witt & J. S. Rovner, eds.). Princeton Univ. Press, Princeton. 440 pp. Maddison, W. P. 1982 (abstract). Stridulation in the agilis group of the jumping spider genus Pellenes. Amer. Arachnol., 26:10. Maddison, W. P. and G. E. Stratton. 1988. Sound production and associated morphology in male jumping spiders of the Habronattus agilis species group (Araneae, Salticidae). J. Arachnol, 16:199-211. Wayne P. Maddison, Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138 USA; Gail E. Stratton, Department of Biology, Bradley University, Peoria, Illinois 61625 USA (present address: Department of Biology, Albion College, Albion, Michigan 49224 USA). Manuscript received January 1987, revised February 1988. A FAUNAL SURVEY OF SPIDERS ASSOCIATED WITH PINUS RADIATA IN A SOUTHERN CALIFORNIA FARM Spiders form an important predatory guild associated with coniferous trees. Their role as predators of lepidopterous pests in such ecosystems has been investigated by several researchers. Eickenbary and Fox (1968) reported spiders as the most abundant predators of the Nantucket pine tip moth (NPTM), Rhyacionia frustrana (Comstock), in loblolly pines, Pinus taeda L., in South Carolina. They also reported that adult NPTM were captured in webs of Front inella communis (Hentz) and Argiope aurantia (Lucas); whereas both NPTM adults and larvae were preyed upon by Metaphidippus galathea (Walckenaer), Misumenops asperatus (Hentz), and Peucetia viridans (Hentz). Bosworth et al. (1970) studied the spiders associated with loblolly pines in Oklahoma. They found NPTM adults trapped in webs of Cyclosa conica (Pallas), Mangora gibberosa (Hentz), Neoscona spp. and Front inella pyramitella (Walckenaer). Juillet (1961) considered spiders the most effective predators of the 1988. The Journal of Arachnology 16:270 European pine shoot moth, Rhyacionia buoliana (Schiff), due to their abundance and the different stages they attacked. Ohmart and Voigt (1981) sampled arthropods in natural and planted Monterey pine stands in California. Based on foliage samples, they reported spiders to comprise 33 percent of the total individuals and the most abundant arthropod group. In California, NPTM is the key insect pest of Monterey pine, Pinus radiata D. Don, grown commercially as Christmas trees. Under southern California conditions, NPTM goes through four generations per season. This complicates management efforts and leads to poor control due to improper timing and misapplications of pesticides. Attempts by Scriven and Luck (1978; 1981) have been made to introduce parasites against this pest. This approach was successful in relatively undisturbed landscape settings. However, reliance on parasites for NPTM control in commercial Christmas tree production may not be feasible. This is due to frequent cultural practices and control measures directed toward other pests which may disrupt the host/ parasite balance. Spiders, therefore, emerge as potentially valuable biological control agents in such high disturbance settings. Their merits lie in their high mobility, broad carnivorous feeding habits, and relatively high reproductive capability. The study reported herein was conducted to determine the relative seasonal abundance and species diversity of spiders in a commercial Christmas tree farm in southern California. The study was conducted in 1986 on a 3-year-old stand of Christmas trees in Grand Terrace, San Bernardino County, California. Average tree height was 83 cm. Ground-associated spiders were monitored with pitfall traps similar to the method of Greenslade (1964). Eighteen traps were placed in the ground, spaced approximately 3.66 m (12 ft) apart. Traps were changed once every three weeks between April 24 and October 23. Samples were taken to the laboratory for determination and quantification. Foliage-associated spiders were sampled from 48 trees, utilizing the beat pan method modified from Bosworth et al. (1971). Sampling was conducted on May 30, July 9, September 19, and October 24. Kaston (1978) was used for familial, generic and, when possible, specific determinations. Calculations were made of richness (total number of families) and abundance (total number of individuals). Additionally, calculations were made on familial diversity through modification of the Shannon-Wiener index of species diversity: H’= - X(n/ N)\og(n/ A7); where ‘V’ equals the number of individuals of a family in the sample and “A” equals the total number of individuals of all families in the sample. Seventeen families, represented by 24 genera, were captured during the study (Table 1). Quantitative measurements of ground- and foliage-associated spiders are shown in Table 2. A larger number of families and a greater abundance of ground-associated spiders were noticed early in the season. As the season progressed, fewer numbers of a lesser amount of families were captured. This trend was reflected by diversity which was highest early in the season, but decreased by approximately 50 percent at the end of the study. The decrease in diversity was likely due to the family Lycosidae which was abundant throughout the season, but dominated in numbers during the latter part. The most commonly encountered genus in that family was the thin-legged wolf spiders, Pardosa. Foliage-associated spiders also were abundant early but declined later in the season. Their diversity also was high at the early part of the study, decreasing by approximately 50 percent as the season progressed. Salticidae was the most 1988. The Journal of Arachnology 16:271 Table 1. — Spiders captured in a southern California Christmas tree farm, 1986. FAMILY SCIENTIFIC NAME OF TAXA Agelenidae Agelenopsis aperta (Gertsch) Amaurobiidae (undetermined) Anyphaenidae Aysha sp. Clubionidae Castianeria sp.; Trachelas deceptus (Banks); Trachelas sp. Dysderidae Dysdera crocata C. L. Koch Gnaphosidae Cessonia classica Chamberlin; Drassyllus sp.; Sergiolus sp.; Zelotes sp. Linyphiidae (undetermined) Lycosidae Alopecosa sp.; Lycosa sp.; Pardosa sp. Oecobiidae Oecobius annulipes Lucas Oxyopidae Oxyopes salticus Hentz; 0. scalar is Hentz Philodromidae Ebo sp. Pholcidae Pholcus phalangioides (Fuesslin); Physocyclus californicus Chamberlin & Gertsch Pisauridae (undetermined) Salticidae Habronattus sp.; Phidippus johnsoni G. & E. Peckham Tetragnathidae Tetragnatha laboriosa Hentz Theridiidae Latrodectus hesperus Chamberlin & Ivie Thomisidae Misumena vatia (Clerck); Tibellus sp.; Xysticus sp. frequently encountered family on the foliage throughout the season. The terms “ground-” and “foliage-associated,” as used in this context, denote the methods by which spiders were captured. They do not necessarily imply specific habitat associations. The latter can be qualified by the facts that salticids were often captured in pitfall traps, and lycosids — especially Pardosa — were occasionally encountered on the foliage. The spider fauna studied in this Christmas tree farm was most abundant and diverse early in the season. Early NPTM generations are considered more damaging than later ones, due to their feeding on young growing tips. It is conceivable that an abundant and diverse spider fauna during that period may result in a significant reduction in NPTM population through predation on several of the life stages consistent with observations by Juillet (1961). This Table 2. — Quantitative measurements of ground- and foliage-associated spiders based on pitfall trap counts and beat pan samples, respectively, in a commercial Christmas tree farm, Grand Terrace, California, 1986. DATE RICHNESS ABUNDANCE DIVERSITY GROUND-ASSOCIATED May 8 10 79 0.7493 May 30 11 111 0.6533 June 20 9 143 0.4834 July 1 1 7 87 0.5132 Aug. 1 6 33 0.5733 Aug. 22 7 63 0.3582 Sept. 12 4 38 0.4169 Oct. 3 6 58 0.3193 Oct. 23 5 46 0.3823 FOLIAGE-ASSOCIATED May 30 6 22 0.6921 July 9 2 7 0.2342 Sept. 12 3 4 0.4522 Oct. 24 3 4 0.3469 1988. The Journal of Arachnology 16:272 percentage of biological control may then be supplemented with selective insecticides (insect growth regulators) to attain the desired degree of suppression. Therefore, careful manipulation of several components is needed to enhance the beneficial spider fauna in the highly-disturbed commercial Christmas tree agroecosystems. LITERATURE CITED Bosworth, A. B., H. G. Raney, R. D. Eickenbary and N. W. Flora. 1970. Nocturnal observations of spiders in Loblolly pines at Haworth, Oklahoma. J. Econ. Entomol., 63:297-298. Bosworth, A. B., H. G. Raney, E. D. Sturgeon, R. D. Morrison and R. D. Eickenbary. 1971. Population trends and location of spiders in loblolly pines, with notes of predation on the Rhyacionia complex (Lepidoptera: Olethreutidae). Ann. Entomol. Soc. America, 64:864-870. Eickenbary, R. D. and R. C. Fox. 1968. Arthropod predators of the Nantucket pine tip moth, Rhyacionia frustrana. Ann. Entomol. Soc. America, 61:1218-1221. Greenslade, P. J. M. 1964. Pitfall trapping as a method of studying populations of Carabidae (Coleoptera). J. Anim. Ecol., 33:301-310. Juillet, J. A. 1961. Observations on arthropod predators of the European pine shoot moth, Rhyacionia buoliana (Schiff) (Lepidoptera: Olethreutidae), in Ontario. Canadian Entomol, 93:195- 198. Kaston, B. J. 1978. How to Know the Spiders. 3rd ed. Wm. C. Brown Co. Publ., Dubuque, 272 pp. Ohmart, C. P. and W. G. Voight. 1981. Arthropod communities in the crowns of the natural and planted stands of Pinus radiata (Monterey pine) in California. Canadian Entomol., 113:673-684. Scriven, G. T. and R. F. Luck. 1978. Natural enemy promises control of Nantucket pine tip moth. California Ag., 32:19-20. Scriven, G. T. and R. F. Luck. 1981. Biological control of an introduced Monterey pine pest. Gold. Gard., 49:9-10. A. D. Ali, Cooperative Extension, Department of Entomology, University of California, Riverside, CA 92521-0314 USA; and Janet S. Hartin, University of California Cooperative Extension, 777 East Rialto Avenue, San Bernardino, CA 92415-0730 USA. Manuscript received October 1987, revised February 1988. PREY HANDLING AND FOOD EXTRACTION BY THE TRIANGLE-WEB SPIDER HYPTIOTES CAVATUS (ULOBORIDAE) Triangle-web spiders, Hyptiotes cavatus (Hentz), emerge from egg sacs as second instars, begin constructing prey-capture webs when they enter the third stadium, and mature as sixth instars (Opell 1982). During an earlier laboratory rearing study of developmental rates and web production (Opell 1982), I also collected prey remains and recorded prey handling times. Here, I describe the prey necessary for the maturation of H. cavatus , trace developmental changes in its prey handling times and prey extraction rates, and evaluate the feeding efficiencies of each of its instars. 1988. The Journal of Arachnology 16:273 Table I. — Prey consumption and prey extraction during Hyptiotes cavatus development. INSTAR PREY CONSUMPTION PREY EXTRACTION (mg dry weight) No. Spiders Mean No. Prey/ Spider SD No. Prey Mean Extraction Per Fly Stadium Total 3rd 21 3.9 1.8 72 0.08 0.32 4th 21 2.8 0.7 59 0.16 0.43 5th 21 4.7 1.3 97 0.15 0.64 6th — — — 86 0.16 — Total 14 11.0 2.6 All spiders used in this laboratory study were reared from egg sacs and were individually housed in plastic containers that measured 30 X 16 X 8.5 cm. Wooden dowel rods cemented into each container provided web attachment sites. Spiders were kept at 23-25° C and 85-95% relative humidity and maintained on a 10:14 hour light:dark cycle. I checked these spiders daily and blew one wild type Drosophila melanogaster (both males and females were used) into each web they produced. I recorded the duration of a complete prey wrapping sequence and from this subtracted periods of inactivity and prey transport to obtain actual prey wrapping time, I began timing feeding when a prey’s thick silk swathing became transparent as it absorbed digestive enzymes, checked specimens every 10-15 minutes thereafter, and noted when the spider had discarded its prey. These extracted prey were collected, placed in a vacuum desiccator with desiccant, and stored until the study was completed four and one-half months later, at which time they were pooled by instar and weighed on a Metier® H-31 AR balance. At 6-8 week intervals during this study, three samples of 100 fruit flies each were taken from the stock cultures used to feed the spiders, placed in a clean vial, heat-killed by holding the vial over a steam jet for 5 seconds, spread on filter paper, and placed in a vacuum desiccator. From the mean dry weight of these flies (0.19 mg, range 0.17-0.23 mg), I subtracted the mean dry weight of the prey discarded by spiders of each stadium to obtain prey extraction values. Table 1 summarizes the number of prey consumed during each stadium and the amount of material extracted from each prey. The numbers of prey eaten by males and females are combined because /-tests reveal no significant difference (p > 0.05) between them. Only the mean numbers of flies eaten by fourth and fifth instars differ significantly (p < 0.05) when compared with /-tests. The amount of material spiders extract from flies doubles after the third instar, but shows no increase thereafter (Table 1). The small size of third instars may limit the volume of digestive enzymes they can produce and make available to them only half the potential food of a fruit fly. Although the number of prey consumed by third and fourth instars does not differ significantly, this increased extraction by fourth instars is responsible for their having a 34% greater total prey extraction (mean number of prey consumes X mean extraction) than third instars. The greater number of flies eaten during the fifth stadium results in a further 49% increase in food intake. Table 2 documents a 77% decrease in wrapping time and an 84% decrease in feeding time from third to sixth stadia. The percentage of prey handling time devoted to wrapping drops by half after the third stadium, but remains constant thereafter. Extraction efficiency increases during development, with fourth instars 1988. The Journal of Arachnology 16:274 Table 2. — Developmental changes in Hyptiotes cavatus prey handling. All times are in hours. Extraction values are in mg dry weight. INSTAR WRAPPING TIME FEEDING TIME TOTAL TIME PER FLY mg EXTRACTED PER HOUR FEEDING No. Mean SD No. Mean SD No. Mean SD % Wrapping 3rd 31 0.44 0.22 21 6.55 4.31 11 6.25 4.54 7 0.012 4th 35 0.34 0.12 10 10.31 4.18 10 10.67 4.25 3 0.016 5th 42 0.22 0.06 20 5.21 1.42 20 5.41 1.42 4 0.029 6th 20 0.13 0.05 12 3.06 0.70 12 3.18 0.73 4 0.052 acquiring 1.3 times more food per hour of feeding than third instars and fifth and sixth instars each removing 1.8 times more food per hour than subsequent instars (Table 2). The laborious prey wrapping characteristic of uloborids (see Lubin 1986 for a review) may compensate for their lack of poison glands and their inability to inject prey. Lubin (1986) found that prey type and mass influence the thoroughness of uloborid wrapping. All spiders of this study were fed the same type of prey and, judging by the opacity and smoothness of the wrapped flies, wrapping thoroughness remains relatively unchanged during development. Therefore, the shorter wrapping times characteristic of later instars probably reflect increases in the aciniform silk glands and spigots used in prey wrapping (Foelix 1982) and reductions in the time required for spiders to circumscribe a prey during early wrapping stages and to manipulate a partially swathed fly during latter wrapping stages. Together with the previous developmental study of H. cavatus (Opell 1982), these results emphasize the small cost of web production. Accidental web damage caused spiders to receive an average of only 0.84 flies per web constructed. Despite this, their development times did not differ markedly from those of natural populations. Although wrapping and feeding times differ among instars, the proportion of each stadium’s “web construction” phase (Opell 1982) devoted to prey handling remains surprisingly small and constant. Third instars devote 4.8% of this time to prey handling, fourth instars 7.3%, and fifth instars 5.5%. Matthew H. Greenstone and Yael D. Lubin made useful comments on this manuscript. , LITERATURE CITED Foelix, R. F. 1982. Biology of Spiders. Harvard Univ. Press, Cambridge, Massachusetts, 306 pp. Lubin, Y. D. 1986. Web function and prey capture behavior in Uloboridae. Pp. 132-171, In Spider- Webs, Behavior, and Evolution. (W. A. Shear, ed.). Stanford Univ. Press, Stanford. Opell, B. D. 1982. Post-hatching development and web production of Hyptiotes cavatus (Hentz) (Araneae). J. Arachnol., 12:105-114. Brent D. Opell, Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061 USA. Manuscript received December 1987, revised March 1988. 1988. The Journal of Arachnology 16:275 SPIDER PREDATORS OF MOSQUITO LARVAE Spiders have been largely overlooked as predators of mosquito larvae in aquatic ecosystems. Bishop and Hart (1931) were the first to report a spider {Pardos a sternalis (Thorell)) consuming mosquito larvae in a small gravel pit pool in Colorado. Garcia and Schlinger (1972) also reported consumption of mosquito larvae by P. sternalis . The mosquitos involved in the latter instance were Aedes dorsalis (Meigen) breeding in a California salt marsh. Similarly, Greenstone (1979, 1983) reported evidence of predation of Ae. dorsalis by Pardosa ramulosa (McCook). Dolomedes sp. was found to prey upon 32P-labeled Culex pipiens pipiens L. larvae in a Southeast Texas ricefield (Breene, unpubl. data). Finally, Service (1973) found a species of Lycosa and one of Pardosa testing positive for Anopheles gambiae Giles in a precipitin analysis, but implied they probably attacked only emerging mosquito adults. In the current study, a pisaurid, Dolomedes triton (Walckenaer), and two lycosids, Pirata sedentarius Montgomery and Pardosa delicatula Gertsch & Wallace, were evaluated as predators of C. p. pipiens , the northern house mosquito. Dolomedes triton and P. sedentarius were chosen for their close association with mosquito larvae habitats in East Texas, while P. delicatula was chosen due to its common presence in grassy areas that border much of the mosquito larvae habitat in the College Station, Texas area. The results of these evaluations are reported within. METHODS Fourth instar C. p. pipiens larvae from laboratory cultures were irradiated with 0.1 to 0.4 juCi/ml 32P for 24 h in a 500 ml container, and then were removed and washed thoroughly to remove residual radioactivity from the integument. A mean DPM (disintegrations per minute) for the mosquito larvae («=50) was determined from a random sample of larvae before each experiment. Approximately 1000 radioactive larvae were placed into each of two simulated grass bank ponds, each measuring 110 cm by 70 cm and filled with water to a depth of 8 cm. These ponds were set up in aquarium tanks lined with black plastic tarp to facilitate rapid removal of any radioactive residues between experiments. Approximately 25% of the surface of the water in each pond was covered with duckweed ( Wolffia papulifera Thomps. and Spirodela ologorhiza (Kurtz) Hegelm) and grass debris {Cynodon dactylon (L.)) to simulate natural pond conditions. An additional 1000 non-radioactive mosquito larvae were placed in an identical control tank. The first simulated pond contained only spiders captured on or near local ponds. The second pond contained both spiders and several of the 30 species of aquatic insect predators also found in local ponds that were used over the course of the experiments. In the case of the control, 1000 non-radiated mosquito larvae were placed in a simulated pond containing both spiders and insects. Otherwise, the control pond was similar in all aspects to the test ponds. After 48 h, all spiders and insects were removed from the test and control ponds and subjected 1988. The Journal of Arachnology 16:276 individually to liquid scintillation counting procedures. Seventeen replications were performed. A simple linear algorithm was used to estimate quantitative ingestion of larvae by the three species of spiders. Observation of predation of a known number of mosquito larvae with a known radioactive mean by each species of spider was used to derive the quantifying algorithm. A complete and detailed account of 32P quantitative methods can be found in Breeee & Sterling (1988). RESULTS AND DISCUSSION A total of 56 of 73 (76.7%) D. triton exposed to y P- labeled mosquito larvae were found labeled with 32 R An average of 12 mosquito larvae per 24 h were consumed by these labeled spiders. Only six D. triton used in the study were adult, of which half were radioactive, indicating larval consumption. In the case of P sedentarius , 118 of 160 (73.8%) of the spiders consumed an average of two mosquito larvae per 24 h. However, only 17 of 56 (30.4%) R delicatula tested positive for radioactivity. Of the Pirata and Pardosa utilized, 106 of 160, and 51 of 56 were adults, respectively. No significant differences were found in predation rates between any of the spider species in either the tank with spiders only or in the tank where the spiders were given a wider choice of prey in the form of other insects. Both Dolomedes and Pirata were observed preying upon the mosquito larvae by grasping individual larvae from beneath the surface of the water, pulling their bodies through the surface tension and consuming them. Dolomedes triton and P sedentarius share habitat preferences in common with mosquito larvae (Carico 1973; Wallace and Exline 1978; Heiss and Meisch 1985). In Texas, these spiders most notably associate with riceland populations of Psorophora columbiae (Dyar and Knab) and a salt marsh mosquito, Aedes solicitans (Walker). Pardosa delicatula is often found along the banks of ponds and streams but has not been closely tied with the aquatic habitat. However, other species of Pardosa have been found in such habitats (Bishop and Hart 1931; Garcia and Schlinger 1972; Greenstone 1979, 1980; Heiss and Meisch 1985). Dolomedes triton and many species of Pirata are found commonly associated with mosquito larva habitats except during reproductive or migration a! cycles. In salt marshes, hunting spiders such as Pirata (LaSalle and Cruz 1985) and Dolomedes (pers. obs.) may be highly important invertebrate predators of mosquito larvae due to the paucity of freshwater aquatic insect predators known to prey upon mosquito larvae. This study furnishes laboratory evidence that the three species of spiders tested will prey readily upon mosquito larvae. If a complete picture of the predation ecology of culicine larvae is to be ascertained, field work that includes entire groups of potentially important tax a, such as the Araneae, will be required. Sincere appreciation is extended to C. D. Dondale for identification of the lycosids and to C. Burandt for identification of the botanical specimens. Many thanks go to D. A. Dean, P. M. Langan, J. Langan and S. Stewart for helpful suggestions involving the manuscript. Portions of this study were supported in part by grants from the U.S. Environmental Protection Agency and the USDA Special Grants Office (Grant No. CR806771-03), 1988. The Journal of Arachnology 16:277 LITERATURE CITED Bishop, S. C. and R. C. Hart. 1931. Notes on some natural enemies of the mosquito in Colorado. J. New York Entomol. Soc., 39:151-157. Breene, R. G. and W. L. Sterling. 1988. Quantitative phosphorus-32 labeling method for predator analysis of the cotton fleahopper, Pseudatomoscelis seriatus (Hemiptera: Miridae). J. Econ. Entomol. In press. Carico, J. E. 1973. The nearctic species of the genus Dolomedes (Araneae: Pisauridae). Bull. Mus. Comp. Zook, 144:435-488. Garcia, R. and E. I. Schlinger. 1972. Studies of spider predation on Aedes dorsalis (Meigen) in a salt marsh. California Mosq. Cont. Assoc. Proc., 40:117-118. Greenstone, M. H. 1979. Spider feeding behavior optimises dietary essential amino acid composition. Nature, 282:501-503. Greenstone, M. H. 1980. Contiguous allotopy of Pardosa ramulosa and Pardosa tuoha (Araneae: Lycosidae) in the San Francisco Bay Region, and its implications for patterns of resource partitioning in the genus. American Midi. Nat., 104:305-311. Greenstone, M. H. 1983. Site-specificity and site tenacity in a wolf spider: A serological dietary analysis. Oecologia, 56:79-83. Heiss, J. S. and M. V. Meisch. 1985. Spiders associated with rice in Arkansas with notes on species compositions of populations. Southwest. Nat., 30:119-127. LaSalle, M. W. and A. A. de la Cruz. 1985. Seasonal abundance and diversity of spiders in two intertidal marsh plant communities. Estuaries, 8:381-393. Service, M. W. 1973. Mortalities of the larvae of the Anopheles gambiae Giles complex and detection of predators by the precipitin test. Bull. Entomol. Res., 62:359-369. Wallace, H. K. and H. Exline. 1978. Spiders of the genus Pirata in North America, Central America and the West Indies (Araneae: Lycosidae). J. Arachnol., 5:1-112. R. G. Breene, M. H. Sweet and J. K. Olson, Department of Entomology, Department of Biology, and Department of Entomology, Texas A&M University, College Station, Texas 77843 USA. Manuscript received June 1987, revised August 1987. 1988. The Journal of Aracfanology 16:278 REGINALD FREDERICK LAWRENCE, 1897-1987 Dr. Reginald Frederick Lawrence, dean of African Aracheology, died in Pietermaritzburg, South Africa on October 9, 1987 at the age of 90, after a brief illness. He left behind a legacy of contributions to science in general and Arachnology in particular which spanned more than 60 years. Dr. Lawrence was born in the small coastal town of George in the Cape Province of South Africa on March 6, 1897. He was educated from 1908 to 1913 at Saint Andrew’s College in Grahamstown, and he matriculated from Tulbagh High School in 1915. He went on to study at the University of Cape Town (then the South African. College), but his studies were interrupted by World War I. He spent two years as an infantryman in France, being wounded in 1918. After recovering from his wounds, he returned to his University studies and graduated with Ms R.Sc. in 1922. In 1922 he joined the Staff of the South African Museum in Cape Town. At that time his knowledge of Arachnids was minimal, and the then director of the Museum, Dr. Peringuey, hurled the two huge volum.es of Simon’s Histoire Naturelle des Araignies at him and ordered him to absorb the contents if he wanted a job. Borrowing a French dictionary, he succeeded in this task and was appointed on probationary status as assistant in charge of Arachnid a., Myriopoda, Reptilia, and Amphibia. The appointment was subsequently made permanent, and he remained at the South African Museum until 1935. During his early years at the South African Museum he began the extensive course of fieldwork that was to mark his entire career. His first collecting expedition in 1923 was to Mozambique, where he traveled alone, much of the time by donkey- back, along the undeveloped coast. For three months each during 1923/1924/. 1925 he journeyed north Into South West Africa, first by rail to the northern part of the territory, and thee via ox or donkey wagon through Gvamboland, the Kaokoveld, and to the Angolan border. Lawrence was the only expedition member who could shoot, and his fellow expedition members depended on him to fill the pot with fresh meat, usually springbok, which were then present in many thousands. The extensive collection of Arachnida made during these trips formed the basis for his doctoral thesis, for which he received his PhD from the University of Cape Town in 1928. In 1935 Dr. Lawrence was appointed Director of the Natal Museum in Pietermaritzburg, where he remained until his retirement in 1964. He edited the Annals of the Natal Museum from 1935 until 1964. It was during his time at the Natal Museum that he developed his keen Interest In the cryptic fauna of the indigenous forests of southern Africa, culminating in his masterpiece of synthesis, The Biology of the Cryptic Fauna of Forests , published in 1953. He recognized the ancient distributional patterns shown by many of these small animals, and appreciated parallel relationships of the African forest biota to other tropical areas and to other temperate southern continents. His pioneering work on southern African forest biogeography serves as an Inspiration for a new generation of arachnid biogeographers. 1988. The Journal of Arachnology 16:279 Fig. 1. — Dr. R. F. Lawrence in 1984 in Pietermaritzburg, South Africa, on the occasion of this 87th birthday. (Photo by P. M. C. Croeser) Dr. Lawrence was a superb collector, and much of the new material described by him was from his own collections. During his tenure at the Natal Museum he visited indigenous forests from the southern Cape to the Limpopo River, and from the Indian Ocean coast to the crest of the Drakensberg Mountains. During these excursions he was accompanied and assisted by his wife, Ella Thompson Pratt Yule. In addition, he visited and made collections in Madagascar, Mauritius, Mozambique, South West Africa, and Zimbabwe (then southern Rhodesia). The collections amassed by him continue to be a treasure trove of new and exciting taxa, particularly those showing Gondwanan affinities. Dr. Lawrence published 210 scholarly papers and books during a period spanning nearly 60 years. These covered a wide range of topics, including natural history, biogeography, museology, and the taxonomy and biology of Acarina, Araneae, Chilopoda, Diplopoda, Onychophora, Opiliones, Pedipalpi, Pseudoscor- piones, Reptilia, Scorpiones, Solifugae, and Uropygi. His last book. The Centipedes and Millipedes of Southern Africa: a Guide , was published in 1984. He received numerous honors during his career. In 1935 he was elected a fellow of the Royal Society of South Africa; he was elected President of the Entomological Society of Southern Africa in 1953; in 1956 he was awarded the Medal and Grant of the South African Association for the Advancement of Science, and in 1958 was elected President of Section D of that same society; in 1964 the Natal Museum published a Festschrift in his honor; in 1973 he was 1988. The Journal of Arachnology 16:280 awarded the Medal of the Zoological Society of South Africa; in 1985 he was made an honorary member of the American Arachnological Society; and in 1986 he was made an honorary life member of the South African Museums Association. More detailed biographical sketches may be found in the Annals of the Natal Museum , vol. 16, pp. i-ix, 1964; and American Arachnology , vol. 21, pp. 13-15, 1980. Throughout his scientific career, through his retirement, and up until the end of his life, Dr. Lawrence remained a true humanitarian. He was generous, courteous, humble and kind, qualities which he showed to friends and colleagues at all times. Throughout his life he was a solicitous and dedicated correspondent, and spared no effort to be of assistance to established scientists and students alike. Not a letter was received, even from persons that he had never met, that did not receive a careful response. I remember him, at the age of 88, mounting a search in the rugged montane forests of Natal for live specimens of Onychophora which were essential to the doctoral research of a student in Europe. The walking worms were captured alive and duly dispatched via the post to Germany. He was frequently acknowledged for his advice to and efforts on behalf of interested naturalists from around the world. He was a source of support and inspiration to Arachnologists throughout Africa and beyond. Dr. Lawrence leaves behind two sons, Alastair and Jonathan, two sisters, and many friends and colleagues whose privilege and good fortune it was to have known “Lawrie” during his long and productive life. Charles E. Griswold, Department of Entomology, American Museum of Natural History, Central Park West at 79th Street, New York, New York 10024 USA. * B THE AMERICAN ARACHNOLOGICAL SOCIETY President : William A. Shear (1987-1989) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 Membership Secretary : Norman I. Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: James W. Berry (1987-1988) Department of Zoology Butler University Indianapolis, Indiana 46208 Directors: James C. Cokendolpher (1987-1989), W. G. Eberhard (1986-1988), Jerome S. Rovner (1987-1989). Honorary Members: P. Bonnet, W. J. Gertsch, H. Homann, R. F. Lawrencef, H. W. Levi, G. H. Locket, A. F. Millidge, M. Vachon, T. Yaginuma. The American Arachnological Society was founded in August, 1972, to promote the study of Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachnology. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $30.00 for regular members, $20.00 for student members and $50.00 for institutions. Correspondence concerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addition, members receive the bi-annual newsletter of the Society, American Arachnology. American Arachnology, edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-find literature, information about arachnology courses and professional meetings, abstracts of papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arachnology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. President-Elect: George W. Uetz (1987-1989) Department of Biological Sciences University of Cincinnati Cincinnati, Ohio 45221 Treasurer: Gail E. Stratton (1987-1989) Department of Biology Albion College Albion, Michigan 49224 Archivist: Vincent D, Roth Box 136 Portal, Arizona 85632 t Deceased Research Notes Zelotes santos (Gnaphosidae, Araneae): Description of the male from Sierra de la Laguna, B.C.S., Mexico, Maria-Luisa Jimenez 253 Transition from predatory juvenile male to mate-searching adult in the orb-weaving spider Nephila clavipes (Araneae, Araneidae), Leann Myers and Terry Christenson 254 Male residency on juvenile female Nephilia clavipes (Araneae, Araneidae) webs, Elizabeth M. Hill and Terry Christenson 257 Natural history observations of Salticus austinensis (Araneae, Salticidae) in North-Central Texas, Norman V Horner , Frederick B. Stangl, Jr. and G. Kip Fuller 260 Functional aspects of the male palpal organ in Dolomedes tenebrosus , with notes on the mating behavior (Araneae, Pisauridae), Petra Sierwald and Jonathan A. Coddington 262 Anateris festae Borelli, espece de scorpion caracteristique du centre d’endesisme “chimborazo” en equateur, Wilson R. Lourengo 266 A common method of sound production by courting jumping spiders (Araneae, Salticidae), Wayne P. Maddison and Gail E. Stratton 267 A faunal survey of spiders associated with Pinus radiata in a southern California farm, A. D. Ali and Janet S. Hartin 269 Prey handling and food extraction by the triangle-web spider Hypiotes cavatus (Uloboridae), Brent D. Opell 272 Spider predators of mosquito larvae, R. G. Breene, M. H. Sweet and J. K. Olson 275 Obituary Reginald Frederick Lawrence, 1897-1987, Charles E. Griswold 278 CONTENTS THE JOURNAL OF ARACHNOLOGY VOLUME 16 Feature Articles NUMBER 2 Analisis del comportamiento de captura de presas por machos adultos de Metepeira sp. A (Araneae, Araneidae), utilizando telas de juveniles y hembras adultas coespecificos, Carmen Viera y Fernando G. Costa 141 Six new species of Diplocentrus Peters from Central America (Scorpiones, Diplocentridae), Scott A. Stockwell 153 Determinacion de Bryantella speciosa y B. smaragdus , nueva combination, mediante la aplication de tecnicas numericas (Araneae, Salticidae), Cristina Luisa Scioscia 177 Why do “Family Spiders”, Stegodyphus (Eresidae), live in colonies? U. Seibt and W. Wickler 193 Sound production and associated morphology in male jumping spiders of the Habronattus agilis species group (Araneae, Salticidae), Wayne P. Maddison and Gail E. Stratton 199 Ground surface spiders in three central Florida plant communities, David T. Corey and Walter K. Taylor 213 Arboreal spiders (Araneae) on balsam fir and spruces in East- Central Maine, Daniel T. Jennings and John B. Dimond , 223 Morphology of the dorsal integument of ten opilionid species (Arachnida, Opiliones), C. Steven Murphree 237 (continued on back inside cover) Cover photograph, figure on men’s meeting house, Palau, by J. W. Berry Printed by the Texas Tech Press, Lubbock, Texas, U.S.A. Posted at Lubbock, Texas October 7, 1988 The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 16 FALL 1988 NUMBER 3 THE JOURNAL OF ARACHNOLOGY EDITOR: James E. Carico, Lynchburg College ASSOCIATE EDITOR: Jerome S. Rovner, Ohio University EDITORIAL BOARD: J. A. Coddington, National Museum of Natural History, Smithsonian Institution; J. C. Cokendolpher, Texas Tech University; F. A. Coyle, Western Carolina University; C. D. Dondale, Agriculture Canada; W. G. Eberhard, Universidad de Costa Rica; M. E. Galiano, Museo Argentino de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia, Missouri; N. F. Hadley, Arizona State University; N. V. Horner, Midwestern State University; H. W. Levi, Harvard University; E. A. Maury, Museo Argentino de Ciencias Naturales; M. H. Muma, Western New Mexico University; N. I. Platnick, American Museum of Natural History; G. A. Polis, Vanderbilt University; S. E. Riechert, University of Tennessee; A. L. Rypstra, Miami University, Ohio; M. H. Robinson, U.S. National Zoological Park; W. A. Shear, Hampden-Sydney College; G. E. Stratton, Albion College; W. J. Tietjen, Lindenwood College; G. W. Uetz, University of Cincinnati; C. E. Valerio, Universidad de Costa Rica. THE JOURNAL OF ARACHNOLOGY (ISSN 0161-8202) is published in Spring, Summer, and Fall by The American Arachnological Society at Texas Tech Press. Individual subscriptions, which include membership in the Society, are S30.00 for regular members, $20.00 for student members. Institutional subscriptions to The Journal are $50.00. Correspondence concerning subscriptions and member- ships should be addressed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. Susan E. Riechert, Department of Zoology, University of Tennessee, Knoxville, TN 37916 U.S. A., at $15.00 for each number, $40.00 for a complete volume; the Index to Volumes 1-10 is available for $10.00. Remittances should be made payable to The American Arachnological Society. Correspondence concerning undelivered issues should be addressed to the Texas Tech Press, Texas Tech University, Lubbock, Texas 79409 U.S. A. Change of address notices must be sent to the Membership Secretary. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without benefit of review. Manuscripts and all related correspondence must be sent to Dr. Jerome S. Rovner, Associate Editor, Department of Zoological Sciences, Ohio University, Irvine Hall, Athens, Ohio 45701 U.S. A. PROOFS, REPRINTS, and CHARGES: Authors will receive a reprint order form along with their proofs. Reprints are billed at the printer’s current schedule of costs. All authors will be billed by the printer for page charges. The charge per journal page will be assessed as follows: $30. 00-nonmembers; $25. 00-members acknowledging grant, institutional, or other support; $ 15.00-full members who do not have support; $ 10.00-student members who do not coauthor with full members and who do not acknowledge support. Perez-Miles, F. and R. M. Capocasale. 1988. Revision of the genus Pycnothele (Araneae, Nemesiidae). J„ Arachnol., 16:281-293. REVISION OF THE GENUS PYCNOTHELE (ARANEAE, NEMESIIDAE) Fernando Perez-Miles and Roberto M. Capocasale Institute) de Investigaciones Biologicas Clemente Establf A Q IS* Division Zoologia Experimental Ave. Italia 3318, Montevideo, Uruguay ^'B'EARiES ABSTRACT On the basis of a character analysis, the genus Pycnothele and species attributed to Androthelopsis were revised and it was concluded that Pycnothele Chamberlin, 1917 = Androthelopsis Mello-Leitao, 1934. The genus Pycnothele comprises three species that are redescribed and illustrated: Pycnothele auronitens (Keyserling, 1891) {---Androthelopsis modestus : Raven, 1985 (in part.) and Psalistops auripilus Mello-Leitao, 1946 new synonyms); Pycnothele perdita Chamberlin, 1917; and Pycnothele singularis (Mello-Leitao, 1934) new combination (= Pycn o the l op sis modestus Schiapelli & Gerschman, 1942 and Androthelopsis modestus: Raven, 1985 (in part) new synonyms). Heteromma anomala Mello-Leitao, 1934, although it belongs to Pycnothele , is an uncertain species. A taxonomic key is included for species identification. INTRODUCTION Pycnothele was created by Chamberlin (1917) based on the type of Pycnothele perdita , from Brazil. This genus includes medium-sized species (usually 20 to 30 mm in body length) found only in South America (Argentina, Brazil and Uruguay). Schiapelli & Gerschman (1942) created the genus Pycnothelopsis and placed It together with Pycnothele in the family Pycnothelidae. These taxa were revised by Mello-Leitao (1934, 1946), Schiapelli & Gerschman (1942), Schiapelli & G. de Pikelin (1965, 1967, 1971), Gerschman de Pikelin & Schiapelli (1970), Capocasale & Perez-Miles (1979), Perez-Miles & Capocasale (1982, 1983) and Raven (1985). Recently Raven (1985) has analyzed the infraorder Mygalomorphae, clarifying the relationships of families. As a result of this analysis, Pycnothele and Androthelopsis were placed in the family Nemesiidae and Pycnothelopsis was designated as a junior synonym of Androthelopsis. The repeated changes of place of the species of these genera and controversies among the authors reveal uncertainty about the correct placing of such taxa and their systematic relations. The diagnostic characters separating Androthelopsis and Pycnothele , apparently clear in the literature, appear to us to be inaccurate or conflicting. Doubtless, the low number of species in these genera has contributed to maintaining restrictive diagnostic criteria for them, a practice which we feel is unjustified. The small number of available specimens has also made the study of intraspecific variation difficult and prejudiced the specific diagnoses and identification. 282 THE JOURNAL OF ARACHNOLOGY As a result of the character analysis we have made on all material available in collections, (1) the species attributed to these genera are distinguished and characterized and (2) their systematic relations are clarified. The election of Pycnothele and Androthelopsis as a unit for study is based on the systematic proximity of these genera, which have traditionally been linked and are now considered sister groups (Raven 1985:45). A key conclusion of our present study is that Pycnothele = Androthelopsis. METHODS All drawings were made with a camera lucida and the measurements with an ocular micrometer; carapace measurements are accurate to 0.1 mm, eye and bulb measurements to 0.025 mm. Computer programs were the Presta package developed in the Centro Ramon y Cajal, Espana. In the Student’s t- test, the confidence limit was R^O.OS; in the correlation calculation, the confidence limit was 95%. In the analysis of character polarity (group under study: species attributed to Pycnothele and Androthelop- sis), Neodiplothele Mello-Leitao was used as out-group. The selection of the out- group was based on the results given by Raven (1985:45). Abbreviations. — British Museum (Natural History), London, England (BMNH); Institute Butantan, Sao Paulo, Brazil (IB); Museo Argentine de Ciencias Naturales Bernardino Rivadavia, Buenos Aires, Argentine (MACN); Museum of Comparative Zoology, Harvard University, Cambridge, USA (MCZ); Museo Nacional de Historia Natural, Montevideo, Uruguay (MNHN). AME= anterior median eyes; ALE= anterior lateral eyes; PME“ posterior median eyes; PLE= posterior lateral eyes. CHARACTER ANALYSIS Integral / pseudosegmented tarsi. — This character was introduced into the systematics of Pycnothelinae by Raven (1985:11). The criteria used to define the pseudosegmented character state were . . tarsi have either a ventral transverse suture (“cracked”), or the cuticle of the lower surface is pallid and has shattered appearance like drying mud (“pseudosegmented”). Pseudosegmented tarsi appear either bent or curved”. The definition of this character is considered undesirable because it implies, at least, three attributes reduced to a double state character. Each attribute could vary independently and no homologies can be established among them. We think it is correct to develop it into three characters: cracked/ not cracked; curved / not curved; and pallid / not pallid. Raven (1985:100) used only the integral tarsi of females to distinguish between Pycnothele and Androthelopsis. Such a character is not comparable because females of Androthelopsis are unknown. Raven (1985:100 and 101) mentioned tarsi I-II (in males of Pycnothele) and (apparently) I-IV (in Androthelopsis) as being pseudosegmented. We examined the types and did not observe “cracked” tarsi in either Pycnothele or Androthelopsis (Figs. 5 and 6). Species attributed to both genera have tarsi lightly curved ( Figs. 2, 4, 6). The pallid condition of the tarsi shows intraspecific PEREZ-MILES & CAPOCASALE— REVISION OF PYCNOTHELE 283 1 2 Figs. 1-6. — Tarsi IV from holotype males of species of Pycnothele: 1, 3, 5, ventral views; 2, 4, 6, lateral views; 1, 2, P. auronitens (= P auripilus = A. modestus [in part]) 3, 4, P. perdita\ 5, 6, P singularis (= A. singularis = A. modestus [in part]). Arrows show the stripe of longer setae dividing scopulae (scapulae divided). variation, which also would be an artifact of preservation. The absence of morphological gaps in these characters do not support the separation of genera. Entire / divided scopulae on tarsi IV. — The character entire / divided scopulae, has been traditionally used in the systematics of Mygalomorphae to separate genera and subfamilies (Schiapelli & Gerschman 1973:43). The type of P perdita presents entire scopulae on tarsi IV (Fig. 3). The holotype and other specimens of P. auronitens examined and species attributed to Androthelopsis by Raven (1985:101 and 102) present the scopulae on tarsi IV longitudinally divided by a stripe of longer setae (Figs. 1 and 5). Raven (1985:100) described scopulae on tarsi IV as entire in Pycnothele (males); our results confirm that description for P perdita; however, the type of P. auronitens presents divided scopulae on tarsi IV. There are two ways to interpret this: (1) P. auronitens is either misplaced in Pycnothele or (2) the character lacks diagnostic value. Neodiplothele has divided 284 THE JOURNAL OF ARACHNOLOGY scopulae on tarsi IV (Raven 1985:102); by out-group comparison, divided scopulae constitute a plesiomorphy. By the criterion of ontogenetic precedence, scopulae division disappears during growth in some Mygalomorphae (Theraphosi- dae) (Schiapelli & Gerscfaman 1973:43); this would corroborate such hypothetic polarity. In the group under study, entire scopulae on tarsi IV can be interpreted as an autapomorphy of P perdita . Therefore, this character appears to be diagnostically useless in these genera. It does not support the separation of genera. Palpal bulb morphology. — The species attributed to Pycnothele and Androthe - lopsis present palpal bulbs with similar morphology. They are pyriform with a conspicuous subapical constriction and a short (5-9% of the bulb length) and narrow (5-10% of maximum bulb width) embolus. Bulbs possess subapical wide vanes, considered as a synapomorphy in these taxa (Raven 1985:45). In Pycnothele , Raven (1985:100) describes: “very high vanes” and in Androthelopsis (1985:101): “high vanes”. In the study of the types and other material examined, it was observed that vane height varies directly with bulb size (r = 0.769, p <0.01), and body size (r = 0.755, p <0.05). Bulb size was also correlated with body size (r = 0.984, p <0.01). The difference pointed out for this character does not constitute a morphological discontinuity that permits a clear delimitation of states. Still, if one were to consider the “very high vanes” state as differentiable (ignoring the correlation with bulb size), it would be exclusively applicable to P perdita (which has bulbs of perceptibly greater size than the remaining species studied). This condition could be interpreted as a specific autapomorphy, without importance in the separation of genera. Consequently, this character appears to be of no value in separating Pycnothele from Androthelopsis. The species attributed to Pycnothele and Androthelopsis share the presence of a well differentiated embolus, short and narrow, that can be distinguished from the rest of the Pycnothelieae. In bulbs having a well differentiated embolus, the short embolus has been considered as plesiomorphic of the Nemesiidae (Raven 1985:80). In the Pycnothelinae, excepting the group submitted to study, the rest of the genera have bulbs with the embolus little or not differentiated (. Neodiplothele , Rachias , Pselligmus ); or differentiated and long (Chaco). According to Raven, (1985:45) Neodiplothele is a sister genus of Pycnothele plus Androthelopsis / Chaco is a sister genus of these three. These facts question the mentioned polarity for embolus characters. However, the data are too fragmentary to reach any conclusion. Omitting the polarity of such characters and analyzing them in terms of similarity, embolus morphology becomes useful to distinguish the species attributed to Pycnothele and Androthelopsis from the rest of the Pycnothelinae. In the Mygalomorphae, the bulbal duct (“spermophor”) is sclerotized in part of its length and it often appears fused with the exterior bulb wall (Kraus 1984:377). This secures the stability of such a structure for its use as a systematic character. The tract of the bulbal duct can be directly observed through the bulb cuticle in Pycnothelinae. P. auronitens (=P auripilus) has the subterminal part of the duct strongly curved, in the proximal sense (Figs. 10, 13). This character differs perceptibly from that of other species studied (Figs. 11-15). It is not possible to determine the polarity of these bulbal duct character states due to absence of data in the out-group; however, they clearly distinguish P. singularis (=P modesius) PEP.EZ MILES & CAPOCASALE— REVISION OF PYCNOTHELE 285 Figs. 7- 15. — Palpal bulbs from holotype males of species of Pycnothele: 7-9, dorsal views; 10-12, ventral views; 13-15, prolateral views; 7, 10, 13, P. auronitens (= P. auripilus - A. modestus [in part]) (left bulb); 8, 11, 14, P. singularis = A. singularis = A. modestus [in part]) (left bulb); 9, 12, 15, E perdita (right bulb). Shaded area represents visible tract of bulbal duct. and P. auronitens (=P, auripilus). Differences in bulb morphology between the types of P. auronitens and R auripilus were not found. The bulb of the type of P. auripilus is more pallid, possibly due to the use of clearing techniques or to the preservation method. This fact probably made observation of the structures difficult for the previous authors. Except for a little difference in size, other differences in bulb morphology between the types of A. singularis and A. modestus were not found. Bulbal duct tract is considered useful as a specific character. 286 THE JOURNAL OF ARACHNOLOGY Cuspules. — The presence of cuspules on maxillae is shared by all species attributed to Pycnothele and Androthelopsis. Number of maxillary cuspules was used as a diagnostic character between Pycnothele and Pycnothelopsis (sub Androthelopsis) by Schiapelli & G. de Pikelin (1967). Capocasale & Perez-Miles (1979) analyzed the value of this character in Pycnothelopsis , discarding it as generic and specific character because it overlaps with the mentioned values for Pycnothele and because it presents high intraspecific variability. These results were confirmed in the present analysis. Raven (1985:79) considers the presence of maxillary cuspules as a plesiomorphy of Nemesiidae; this criterion agrees with our results in the group submitted to study. The presence of cuspules on the labium is shared by the species of the group under study (Figs. 16=18), except in the type of Heteromma anomala. This character was used by Schiapelli & G. de Pikelin (1967). Capocasale & Perez- Miles (1979) concluded that like the maxillary cuspules, it lacks diagnostic value at the generic or specific level in Pycnothelopsis (sub Androthelopsis ). Raven (1985:100, 101) indicated “No cuspules on labium” in the descriptions of both genera. This statement is only valid for the type of H. anomala but does not have a factual basis for other species studied. It was not possible to establish the polarity of the labial cuspules in the Pycnothelinae. However, Raven (1985:79) indicated that “labium with few cuspules” would represent a plesiomorphy in Nemesiidae. According to the results obtained, we conclude that both maxillary cuspules and labial cuspules do not support the separation of genera. Eyes. — Eye dimensions have been used as diagnostic characters separating Pycnothele and Pycnothelopsis (sub Androthelopsis) by Schiapelli & G. de Pikelin (1967). The correlation analysis between eye dimensions (maximum diameter) and body size (length of carapace) in the specimens of the group under study, gave the following results: AME / carapace r=0.805; ALE /carapace r=0.932; PME / carapace r=0.737; PLE / carapace r=0.854. These values indicate a significant correlation at the 95% level. This leads us to question the systematic value of this character, as it is empirically correlated with size. To avoid variations due to size of specimens, eye dimensions were studied in a relative way (maximum diameter / carapace length). Significant differences in the relative dimensions of eyes between the type of P perditus and the sample of P singulars (=A. singularis = P. modestus) were not found. In the comparison of P perdita with P. auronitens (=P auripilus = A. modestus in part) AME and PME presented significant differences (t= 10.42; P <0.001 and t= 4.02; P <0.02 respectively). Significant differences for ALE and PLE were not found. (In the type of P auronitens the right PLE is ectopic and of lesser size. It was not used in the comparison). In the comparison between the samples of P. auronitens and P. singularis only ALE show significant differences (/=2.04; P <0.05). The two species placed in different genera (P. perdita and P. singularis) by Raven (1985:100, 101) do not show differences in these characters. P. perdita and P, auronitens placed by Raven (1985:100) in the same genus, have differences in the relative size of AME and PME. Taking into account the absence of data that could permit us to determine polarity of these characters, and the results obtained, they are considered to be specific level characters. Such characters do not support the separation of genera. PEREZ-MILES & CAPOCASALE— REVISION OF PYCNOTHELE 287 Figs. 16-18. — Labia and maxillae of male holotype of species of Pycnothele , ventral views: P. auronitens (= P. auripilus - A. modestus [in part]); 17, P. perdital\ 18, P. singularis (= A. singularis - A. modestus [in part]). Arrows show cuspules. Genus Pycnothele (Chamberlin, 1917) Pycnothele Chamberlin, 1917:26; Mello-Leitao 1923:39; Petrunkevitch 1928:73; 1940:303; Berland 1932:329; Gerhardt & Kaestner 1939:591; Neave 1940:1051; Roewer 1942:275; Schiapelli & Gerschman 1942:319; Schiapelli & G. de Pikelin 1965:15; 1967:53; G. de Pikelin & Schiapelli 1970:100; Bonnet 1958:3836; Perez-Miles & Capocasale 1983:2; Raven 1985:100. Trechona: Keyserling 1891:16 (in part). Crypsidromus : Simon 1903:931 (in part); Biicherl 1952:132 (in part). 288 THE JOURNAL OF ARACHNOLOGY Metriopelma : Pocock 1903:112 (in part); Mello-Leitao 1923:168 (in part); Bonnet 1957:2826 (in part). Androthelopsis Mello-Leitao, 1934:402; Roewer 1942:217; Bonnet 1955: 322; Raven 1985:101. NEW SYNONYMY. Heteromma Mello-Leitao, 1935:356 (preoc. by Heteromma Menge 1856, in Neave 1939:640); Bonnet 1957:2184. Agersborgia Strand, 1936:167 (new name for Heteromma ); Bonnet 1955: 205. Pycnothelopsis Schiapelli & Gerschman, 1942:319; Schiapelli & G. de Pikelin 1965:15; 1967:59; Biicherl 1957:408; Capocasale & Perez-Miles 1979:1 (in part); Perez-Miles & Capocasale 1982:1 (in part); 1983:1. Diagnosis. — Pycnothele differs from other Pycnothelinae because the males possess bulbs with differentiated short emboli and subapical wide vanes (Figs. 7- 15); in females, spermathecae each have a long and narrow neck gradually widening apically; fundus subglobulose. DISCUSSION Schiapelli & Gerschman (1942) established the separation between Pycnothele and Pycnothelopsis according to the following characters: scopulae extension on metatarsi I and II, labial and maxillary cuspulae and ocular dimensions. Capocasale & Perez-Miles (1979) and Perez-Miles & Capocasale (1982, 1983) invalidated some characters considered as diagnostic in these genera, although they maintained them as separate taxa. Raven (1985:101) established the synonymy between Pycnothelopsis and Androthelopsis , maintaining the species under study in two separate genera: Pycnothele and Androthelopsis. This author based the separation on the following characters: integral/ pseudosegmented tarsi and height of vanes on bulb. According to the preceeding analysis, the characters considered as diagnostic of Pycnothele and Androthelopsis have no value. The mentioned differences between these genera are either erroneous or do not justify that they be maintained as separate taxa. Proper synapomorphies of each genus that can justify their separate existences as monophyletic groups were not found. Raven (1985:45) indicated that wide vanes on the bulb are a synapomorphy of Pycnothele plus Androthelopsis (sister groups). We agree with this author in the polarity assigned to the character, but we consider that it is a synapomorphy of generic level, which indicates the monophyly of the species attributed to both genera. Using similarity criteria, bulb morpohology and embolus length are more similar among the species attributed to Pycnothele and Androthelopsis than they are between any of these species and the other members of the Pycnothelinae. A significant morphological discontinuity observed among the species attributed to Pycnothele and Androthelopsis involved the character, entire/ divided scopulae of tarsi IV. This character has been traditionally used to separate genera in Mygalomorphae. If only this character is considered, the species would be placed in two genera; (1) Androthelopsis plus P. auronitens , with divided scopula and (2) Pycnothele (monospecific). But since divided scopulae on tarsi IV are plesiomorphic, Androthelopsis plus P. auronitens would constitute a genus based on symplesiomorphy. If Pycnothele remained as a monospecific genus and sister group of Androthelopsis ( sensu Raven 1985), both PEREZ-MILES & CAPOCASALE— REVISION OF PYCNOTHELE 289 taxa would be paraphyletic ( sensu Platnick 1976). Other morphological discontinuities justifying the existence of Pycnothele and Androthelopsis as separated genera were not found. . The results obtained have induced us to establish the synonymy between Pycnothele and Androthelopsis. Pycnothele (valid name for priority ICZN, art. 23) would be based on the following synapomorphy: wide and conspicuous subapical vanes on bulb (Figs. 7-15). KEY TO SPECIES OF THE GENUS PYCNOTHELE Males 1. Scopulae entire on tarsi IV (Fig. 3) P. perdita Scopulae on tarsi IV, divided by a stripe of thicker and longer setae 2 2. - Bulbal duct presenting a strong subterminal curvature basal ly (Figs. 7, 10, 13) P. auronitens Bulbal duct without such curvature (Figs. 8, 11, 14) P. singulars Pycnothele auronitens (Keyserling, 1891) Figs. 1, 2, 7, 10, 13, 16 Trechona auronitens Keyserling, 1891:16. Metriopelma auronitens : Pocock 1903:114; Mello-Leitao 1923:173; Petrunkevitch 1939:279; Bonnet 1957:2826; 1959:4680. Crypsidromus auronitens: Simon 1903:931; Biicherl 1952:132. Psalistops auripilus Mello-Leitao, 1946:8. NEW SYNONYMY. Pycnothelopsis modestus : Schiapelli & G. de Pikelin 1971:61 (in part). Pycnothelopsis auripilus : Capocasale & Perez-Miles 1979:3 (in part); Perez-Miles & Capocasale 1982:1. Pycnothelopsis auronitens : Perez-Miles & Capocasale 1983:2. Androthelopsis modestus: Raven 1985:102 (in part). NEW SYNONYMY. Pycnothele auronitens: G. de Pikelin & Schiapelli 1970:100; Raven 1985:100. Diagnosis. — P auronitens differs from Pycnothele perdita , by the scopula on tarsus IV which is divided by a stripe of longer setae; from P. singularis by the strong proximal curvature of the bulb duct tract (visible in ventral and prolateral views) (Figs. 10-13). Description. — Male (N= 4): Carapace, length: 5. 6-7. 2 mm (mean = 6.28 ± 0.73 SD), width: 4.4-5. 1 mm (mean = 4.75 ± 0.35 SD). Fovea procurved. Chelicerae without rastellum, intercheliceral tumescence present. Ocular tubercle well defined, longer than wide; AME: 0.18-0.25 mm (mean = 0.11 ± 0.03 SD); ALE: 0.20-0.30 mm (mean = 0.26 ± 0.04 SD); PME: 0.15-0.20 mm (mean = 0.18 ± 0.03 SD); PLE: 0.23-0.35 mm (mean = 0.28 ± 0.05 SD). Labium with 3-5 cuspules. Maxillae subrectangular, distal prolateroventral lobe pronounced, proximal prolateroventral lobe with numerous cuspules. Tibial apophysis absent. Tarsi without spines, with two bipectinated claws. Scopulae on tarsi I-III entire, on tarsi IV divided in half by a longitudinal stripe of longer setae. Apical scopulae on metatarsi I and II; III and IV without scopulae. Sternal sigilla marginal. Anterior spinnerets monoarticulated, short; posterior spinnerets triarticulated, apical segment short and domed. Palpal bulb pyriform with 290 THE JOURNAL OF ARACHNOLOGY subapical wide vanes, embolus differentiated, short; duct-tract of bulb presenting a strong subterminal curvature proximally (visible in ventral and prolateral views). Discussion. — This species was placed in Pycnothele by G. de Pikelin & Schiapelli (1970). Psalistops auripilus (Mello-Leitao, 1946), was transferred to Pycnothelopsis by Schiapelli & G. de Pikelin (1971) (not by Capocasale & Perez- Miles (1979), as Raven said (1985:102)) and placed in the synonymy of P. modestus. Capocasale & Perez-Miles (1979) separated this synonymy into two species: Pycnothelopsis auripilus and Pycnothelopsis modestus. Perez-Miles & Capocasale (1983) transferred P. auronitens to Pycnothelopsis establishing the specific synonymy P. auronitens = P. auripilus. Raven (1985) did not accept this synonymy and placed P. auronitens back in Pycnothele and P. auripilus in Androthelopsis. He based the change on the fact that P auronitens shares with Pycnothele: (1) the absence of pseudosegmented tarsi in the male and (2) elevated vanes on the bulb. The first character state is at odds with his own statement that tarsi I and II of male Pycnothele are pseudosegmented. In any case, both characters became useless as a result of the present analysis. In our present study important differences were not found between the types of P auronitens and P modestus. This confirms the specific synonymy established by Perez-Miles & Capocasale (1983). The synonymy established again by Raven (1985) between A. modestus (“ P singularis) and P. auripilus (= P. auronitens) is overturned. These species are distinguished by the characters mentioned in the diagnosis which agree with the results obtained by Capocasale & Perez-Miles (1979). Material examined.— BRAZIL: Rio Grande, Taquara, holotype male of Pycnothele auronitens (BMNH). URUGUAY: Lavalleja, Arequita (C. de Zolessi) 1 male (MNHN); Maldonado, Sierra de las Animas (Perez, Delgado) 1 male (MNHN); Florida, holotype male of Psalistops auripilus (MNHN). Pycnothele perdita Chamberlin, 1917 Figs. 3, 4, 9, 12, 15, 17 Pycnothele perdita Chamberlin, 1917:26; Roewer 1942:275; Bonnet 1958:3836; Schiapelli & Gerschman 1942:319; Schiapelli & G. de Pikelin 1965:15; 1967:54; Perez-Miles & Capocasale 1983:2; Raven 1985:100. Pycnothele perditus (sic): Mello-Leitao 1923:40; Petrunkevitch 1928:73; Schiapelli & G. de Pikelin 1967:48. Diagnosis. — Males of P. perdita differ from other Pycnothele species by their entire scopulae on tarsi IV and by their bulb morphology ( Figs. 9, 12, 15). Description. — Male: Carapace, length: 14.5 mm; width: 12.2 mm. Fovea procurved. Chelicerae without rastellum, intercheliceral tumescence present. Ocular tubercle well defined, longer than wide; AME: 0.75 mm; ALE: 0.50 mm; PME: 0.28 mm; PLE: 0.50 mm. Labium with 3 cuspules (2 visible plus a base). Maxillae subrectangular, distal prolateroventral lobe pronounced, proximal prolateroventral lobe with numerous cuspules. Tibial apophysis absent. Tarsi without spines, with two bipectinated claws. Scopulae on tarsi I-IV entire. Apical scopulae on metatarsi I and II; III and IV without scopulae. Sternal sigilla marginal. Anterior spinnerets monoarticulated, short; posterior spinnerets triarticulated, apical segment short and domed. Palpal bulb pyriform with subapical wide vanes; embolus differentiated, short; duct tract of bulb gently curved in ventral view (Fig. 15). PEREZ-MILES & CAPOCASALE— REVISION OF PYCNOTHELE 291 Female : Carapace length: 17 mm; width: 13 mm; AME: 0.54 mm; ALE: 0.51 mm; PME: 0.20 mm; PLE: 0.51 mm. Labium with 1 cuspele. Scopulae on tarsi I and II entire; III and IV divided. Spermathecae with long and narrow neck, gradually widening apically, fundus subglobulose. Other characters as in male. Material examined.— BRAZIL: Rio Parahyba, holotype male and paratype female (MCZ). Pycnothele singularis (Mello-Leitao, 1934) NEW COMBINATION Figs. 5, 6, 8, 11, 14, 18 Androthelopsis singularis Mello-Leitao, 1934:402; Roewer 1942:217; Bonnet 1955:322; Raven 1985:101. Pycnothelopsis modestus Schiapelli & Gerschman, 1942:319 NEW SYNONYMY; Schiapelli & G. de Pikelin 1965:15; 1967:59; 1971:61 (in part); Biicherl 1957:405; Capocasale & Perez-Miles 1979:4; Perez-Miles & Capocasale 1982:1; 1983:4; Brigeoli 1983:142. Androthelopsis modestus: Raven 1985:102 (in part) NEW SYNONYMY. Diagnosis. — P singularis differs from R perdita, by having the scopulae on tarsi IV divided; and from P auronilens , by the tract of bulb which lacks subterminal curvature (Figs. 8, 11, 14). Description. — Male (N= 6): Carapace, length: 6.1-11.0 mm (mean = 8.23 + 1.75 SD), width: 4. 6-7. 5 mm (mean = 6.43 ± 1.05 SD). Fovea procurved. Chelicerae without rastellum, intercheliceral tumescence present. Ocular tubercle, well defined, longer than wide; AME: 0.18-0.30 mm (mean = 0.24 + 0.04 SD); ALE: 0.23-0.35 mm (mean = 0.30 ± 0.05 SD); PME: 0.15-0.30 mm (mean = 0.22 ± 0.05 SD); PLE: 0.20-0.35 (mean = 0.29 ± 0.06 SD). Labium with 1-4 cuspules. Maxillae subrectangular, distal prolateroventral lobe pronounced, proximal prolateroventral lobe with numerous cuspules. Tibial apophysis absent. Tarsi without spines, with two bipectinated claws. Scopulae on tarsi I-III entire, scopulae on tarsi IV divided by a longitudinal stripe of longer setae. Apical scopulae on metatarsi I and II; III and IV without scopulae. Sternal sigilla marginal. Anterior spinnerets triarticulated; apical segment short, domed. Palpal bulb pyriform with subterminal wide vanes; embolus differentiated, short; duct- tract of bulb gently curved in ventral view (Fig. 11). Discussion. — P. modestus was transferred to the genus Androthelopsis by Raven (1985:101) who maintained it as a different species from A. singularis. In the type comparison, except for slight differences in size, other important differences in the characters studied were not found. This is the basis of the specific synonymy here established. As a result of the generic synonymy ( Pycnothele — Androthelopsis ), the name Pycnothele singularis , must prevail by priority (ICZN, art. 23). The synonymy established between A. modestus (=P singularis) and P. auripilus (=P. auronitens) by Raven (1985:161) is considered incorrect. These species are differentiated by the characters indicated in the diagnosis. Material examined.— BRASIL: SAO PAULO; Alto da Serra, holotype male of Androthelopsis singularis (IB). ARGENTINA: SANTIAGO DEL ESTERO; Colonia Dora (Prosen), holotype male of Pycnothelopsis modestus (MACN); CORDOBA (Mansilla) 1 male (MACN); CHACO; Colonia Benitez, 1 male (MACN); ENTRE RIOS; Parana, 1 male (MACN). URUGUAY: CERRO LARGO; Rio Tacuari (Costa; Perez) 1 male (MNHN); ARTIGAS; Cerro del Zorro (Gudynas; Skuk) 1 male (MNHN); Arroyo de la Invernada, 1 male (MNHN); SALTO, Rio Arapey (Shanon) 2 males (MNHN). 292 THE JOURNAL OF ARACHNOLOGY UNCERTAIN SPECIES OF PYCNOTHELE Heteromma anomala Mello-Leitao, 1935:356. Holotype male from Brazil, Rio de Janeiro (IB) examined. According to the morphology of the bulb, we agree with Raven (1985) who placed this species in Pycnothele (next to P. perdita). However, (1) it has no cuspules on the labium and (2) tarsi IV are absent in the holotype. Mello-Leitao (1934) did not say If the scopulae on tarsi IV are entire or divided. For these reasons, at present, it is not possible to reach a conclusion and it can only be considered as unidentifiable. ACKNOWLEDGMENTS We are very grateful to the following persons for the loan of collection specimens: P. H Hillyard (BMNH), H. W. Levi (MCZ), S. Lucas (IB), E. Maury (MACN) and to Dr. F. Coyle and Dr. R. Raven for the reviews. We also acknowledge the help of Prof. Nelly R de Perez in translation and typing of the draft of this paper and to Dr. G. Hawksley for the suggestions and corrections of the English. LITERATURE CITED Berland, L. 1932. Les Araehnides, Encyclop. entom., Lechevalier, Paris. 16:1-485. Bonnet, P. 1955. Bibliographia Ara neoram. Toulouse, 2 pt. 1:1-918. Bonnet, P. 1957. Bibliographia Araneorum. Toulouse, 2 pt. 3:1927-3026. Bonnet, P. 1958. Bibliographia Araneorum. Toulouse, 2 pt. 4:3027-4230. Brignoli, P. M. 1983. A Catalogue of the Araneae Described Between 1940 and 1981. Manchester University Press, Manchester, 755 pp. Biicherh W. 1952. Araehas do Rio Grande do Sul. Mem. Inst. Butantan, 24(2): 127-156. Biicherl, W. 1957. Sobre a importancia dos bulbos copuladores e das apofisis tibiais dos machos na sistematica das aranhas caranguejeiras (On ho gnat ha), An. Acad. Bras. Cien., 29:377-416. Capocasale, R. M., & F. Perez-Miles. 1979. Aranas del Uruguay, II. Nuevos aportes al genero Pycnothelopsis Schiapelli y Gerschman, 1942 (Araneae, Pychothelidae). Com. Zool. Mus. Hist. Nat. Montevideo, 10(141): 1-9. Chamberlin, R. V. 1917. New spiders of the Family Aviculariidae. Bull. Mus. Comp. Zool, Harvard University, 61(3):25-75, Gerhard t, U. and A. Kaestner. 1938. Araneae. In: Handbuch der Zoologie. Kiikenthal & Krumbach, Berlin, Leipzig, 3(2):497-656. Gerschman de Pikelin, B. S. y R. D. Schiapelli. 1970. Position sistematica de Metriopelma auronitens (Keys. 1891) (Araneae: Mygalomorphae). Rev. Soc. Ent. Argentina, 32(l-4):99-102. Gerschman de Pikelin, B. S. y R. D. Schiapelli. 1973, La subfamilia “Ischnocolinae” (Araneae: Theraphosidae). Rev. Mus. Cienc. Nat. Buenos Aires, 4(2):43-77. International Commission on Zoological Nomenclature. 1985. International Code of Zoological Nomenclature, 3rd ed., International Trust for Zoological Nomenclature, London, 338 pp. Keyserling, G. E. 1891. Die Spineee Amerikas. Brasilianische Spinnen, Band 3, Verlag von Bauer & Raspe, Nurenberg, 278 pp. Kraus, G. 1984. Male spider genitalia: Evolutionary changes in structure and function. Verb, naturwiss. Ver. Hamburg, 27:373-382. Mello-Leitao, C. 1923. Theraphosoideas do Brasil. Rev. Mus. Paulista 13:1-435. Mello-Leitao, C. 1934. Tres araehas novas nas colleges do lestituto Butantan. Mem. Inst. Butantan, 8:401-407. Mello-Leitao, C. 1935. Tres novas aranhas tetrapneumoees nas colleges do Institute Butantan. Mem. Inst. Butantan, 9:355-360. Mello-Leitao, C. 1946. Nuevos aracnidos sudamericanos para la coieccion del Museo de Historia Natural de Montevideo. Com. Zool. Mus. Hist. Nat. Montevideo, 2(35): 1-10. PEREZ-MILES & CAPOCASALE— REVISION OF PYCNOTHELE 293 Neave, S. A. 1939. Nomenclator Zooiogicus. The Zoological Society of London. London, 2:1025. Perez-Miles, F. y R. M. Capocasale. 1982. Aranas del Uruguay, IV. Hallazgo de una tercera especie del geeero Pycnothelopsis: Pycnothelopsis tacuariensis sp. nov. (Araneae, Pycnothelidae). Com. Zool. Mus. Hist. Nat. Montevideo, 1 1(147): 1-7. Perez-Miles, F. y R. M. Capocasale. 1983. Aranas del Uruguay, VI. Las especies del genera Pycnothele: Pycnothelopsis auronitens (Keyserling, 1891) comb. nov. (Araneae, Pycnothelidae). Aracnologia, 1:1-4. Petrunkevitch, A. 1928. Systema Araeearum. Trans. Connecticut Acad. Arts. Sc., 29:1-270. Petrunkevitch, A. 1940. Catalogue of American Spiders. Trans. Connecticut Acad. Sc., 33:133-338. Platnick, N. I. 1976. Are monotypic genera possible? Syst. Zool., 26(2): 198-199. Poeock, R. I. 1903. On some, genera and species of South American Aviculariidae. Ann. Mag. Nat. Hist, (7)11:81-115. Raven R. J. 1985. The spider infraorder Mygalomorphae (Araneae): Cladistics and systematics. Bull. Amer. Mus. Nat. Hist., 182(1): 1-180. Roewer, C. F. 1942. Katalog der Araneae, Band 1, Verlag von "Natura”, Bremen, 1040 pp, Schiapelli, R. y B. Gerschman. 1942. Aranas argeetinas (1). Anal Mus. Argentina Cs. Nat., 40:317- 331. Schiapelli, R. D., y B. Gerschman de Pikelin. 1965. Distribution de las aranas Mygalomorphae en la Argentina. Act. II. Congr. Lationoamericano Zool, Sao Paulo, 1962, 2:11-20. Schiapelli, R. D. y B. S. Gerschman de Pikelin. 1967. La familia Pycnothelidae (Chamberlin, 1917) (Araneae-Mygalomorphae). Seg. Jorn. Eetomoepid. Argentina, 1:45-64. Schiapelli, R. D. y B. S. Gerschman de Pikelin. 1971. Estudio de algunas aranas descriptas por Mello- Leitao para el Uruguay. Rev. Soc. Ent. Argentina, 33(l-4):57-62. Simon, E. 1903. Histoire naturelle des Araignees 2(4):669-1080, Roret, Paris. Strand, E. 1936. Miscellanea nomenclatoria zoologica et palaeontologica IX. Folia Zool. Hydrobiologia, 4:133-147. Manuscript received May 1987 , revised September 1987. ' ■ Eberhard, W. G. 1988. Behavioral flexibility in orb web construction: Effects of supplies in different silk glands and spider size and weight. J. Arachnol., 16:295-302. BEHAVIORAL FLEXIBILITY IN ORB WEB CONSTRUCTION: EFFECTS OF SUPPLIES IN DIFFERENT SILK GLANDS AND SPIDER SIZE AND WEIGHT William G. Eberhard Smithsonian Tropical Research Institute and Escuela de Biologia, Ueiversidad de Costa Rica Ciudad Universitaria, Costa Rica ABSTRACT Comparisons of webs spun in the field when boih sticky and non-sticky silk supplies were complete, when both were recently depleted, and when only non-sticky supplies were depleted show that Leucauge rnariana and Micrathena sexspinosa vary design features of their orbs such as numbers of radii and sticky spiral loops, web area and proportion covered with sticky spiral, sticky spiral symmetry, and spaces between sticky spiral loops in response to changes in the amounts of both sticky and non-sticky silk that they have available. Spider size, spider weight and possibly website also influence L. Mariana web designs. INTRODUCTION The orbs of araneid spiders are composed of a sticky spiral, produced by the aggregate and flagelliform glands, and a network of non-sticky supporting lines (radii, frames, hub and temporary spiral) drawn from the ampullate glands (Kovoor 1977; Kavanagh and Tillinghast 1979). Several aspects of orb web design have been thought to be species- or genus-specific (Savory 1952; Witt and Baum 1960; Reed and Jones 1965; Witt et al. 1968; Risch 1977; Foelix 1982: Ramousse and LeGueite 1984; Tyshchenko 1984). Individual spiders appear however, to adjust some web characteristics on the basis of their silk supply. Laboratory experiments using drags that stimulated non-sticky silk production altered orb sizes and designs (Witt et al 1968), as did manipulation of non-sticky silk production by altering the spider’s demands for non-sticky silk (Reed et al. 1970). These studies did not take into account, however, other possible effects of the drugs, or possible effects of changes in silk supplies in other glands. The present report of field observations of the araneids Leucauge mariana (Keyserling) and Micrathena sexspinosa (Hahn) shows that recent expenditure of sticky as well as non-sticky silk influences web designs. Spider size, spider weight and possibly websites are also shown to influence web design in L. mariana. MATERIALS AND METHODS Individual spiders were followed during the course of a day by marking websites rather than spiders, as spiders were generally several meters from each 296 THE JOURNAL OF ARACHNOLOGY other, and usually did not change sites during the day (spiders that did move were not included). L. mariana was observed in second growth near San Antonio de Escazu, San Jose Province, Costa Rica, and M. sexspinosa at the edge of a large clearing at the La Selva field station near Puerto Viejo de Sarapiqui, Heredia Province, Costa Rica. All individuals observed were mature females. Webs were measured in the field, and then collapsed by cutting all but three radii near their outer ends with a scissors, leaving all silk still in the web, and the frame and anchor lines intact. Web area was estimated by multiplying vertical length times horizontal width measured from frame to frame. By weighing paper cutouts of 53 photographed L. mariana webs, it was determined that such estimates correlate strongly with area (r = 0.80). The number of sticky loops was the average of the number above and below the hub in M. sexspinosa , and the average of those above, below, and to the right and left of the hub in the less symmetrical webs of L. mariana. The average space between sticky spiral loops was the distance from the inner to the outer loop divided by the number of loops. Those L. mariana spiders which were to be weighed were placed in individual plastic vials with fresh leaves within 1-2 hours of finishing their first webs of the day, and weighed to the nearest 0.1 mg less than 12 hours later on an electrical balance. Using the average rate of weight loss for nine spiders kept in vials for seven hours (0.05 mg/h — spider weights averaged about 45 mg), each spider’s estimated weight when it spun its web was calculated. Each spider was placed in alcohol after being weighed, and cephalothorax and tibia I lengths were measured later using a dissecting microscope. Website effects on orb design were investigated by measuring the first webs of the day for a series of spiders. These spiders were then removed (with little or no damage to the webs) and were replaced the same morning by other spiders taken from finished orbs of their own; the first webs made by the replacement spiders at the same sites were measured the next morning. When possible, statistical tests were performed comparing different webs spun by the same spider on the same day and at the same site, allowing the spider to act as its own control. RESULTS Normal webs. — Both L. mariana and M. sexspinosa replaced damaged webs 1.5-10 hours after their original webs were destroyed at 0700-0900 hours. The second, replacement webs of both species were consistently smaller in area, had fewer radii, and fewer loops of sticky spiral (Table 1). The average spaces between sticky spiral loops were unchanged. Third webs of L. mariana spun the same day to replace destroyed second webs (all third webs were built <15 hours after the first web) showed further reductions in numbers of radii and loops, increased spaces between loops, but no further change in web area (Table 1). The relative portion of L. mariana webs within the outermost loop of sticky spiral also varied. The distances from the outermost sticky loops to the outer ends of the radii were greatest in third, smaller in second, and smallest in first webs (Fig. 1) (first and second webs differed comparing numbers of distances both <0.6 cm and >1.5 cm with Chi Square — p < 0.05, p < 0.001 respectively); second and third differed comparing both <1.1 and >2.0 cm — both p < 0.001 with Chi Square). EBERHARD— FLEXIBILITY IN ORB DESIGN 297 Table 1. — Averages and standard deviations of characteristics of normal first, second and third webs and experimental second webs, and statistical comparisons of ratios from webs spun by the same spiders on the same days at the same sites (two sets of first webs are used in the comparisons of L. mariana webs). Differences between values followed by the same letter are highly significant (p < 0.001) with Mann-Whitney U- test. Web Number of Radii Number of Sticky Loops Area (cm)2 Space Between Loops (cm) N Micrathena sexspinosa First 38.3 + 7.1 25.0 + 6.5 367 + 190 0.23 + 0.04 121 Second 26.2 + 4.4 19.3 + 4.6 286 + 101 0.24 + 0.04 50 Experimental 32.0 + 4.6 24.3 + 4.5 416 + 148 0.25 + 0.06 37 Second/ First 0.66 + 0.16a 0.75 + 24 0.80 + 31b 1.06 + 0.15 45 Exptl. / First 0.83 + 0.09a 1.09 + 0.31b 35 Leucauge mariana First 30.2 + 3.3 41.2 + 10.2 670 + 222 0.23 + 0.03 80 Second 23.4 + 2.7 27.4 + 4.9 453 ± 250 0.24 + 0.04 78 Third 20.7 + 3.0 22.3 + 4.5 403 + 150 0.26 + 0.07 40 Experimental 26.2 + 3.4 35.5 + 7.2 549 + 255 0.22 + 0.04 29 Second / First 0.79 + 0.11c 0.70 + 0. 14d 0.64 ±0.18 1.01 + 0.1 3e 40 Third / First 0.62 + 0.12c 0.53 + 0.1 4d 0.65 + 0.20 1.24 + 0.25e 40 Second / First 0.82 + 0.08f 0.67 + 0.09 0.68 + 0.1 9g 1.05 + 0.14 29 Exptl. / First 0.92 + 0.07f 0.96 + 0.28g 29 Sticky spiral asymmetry was reduced in second and third L. mariana orbs. Absolute values of differences between the average number of loops and the numbers of loops in the four sectors were summed for each web in 44 cases in which the same spider made three successive webs at the same site on the same day. The sums for first webs (x — 8.4 ± 5.0) were larger than those in second (x = 5.0 ± 4.6) and third (x — 4.4 ± 4.2) webs (both p < 0.001 with Mann-Whitney U- test). In addition, when the difference between the maximum and minimum number of loops in each web were compared between first, second and third webs, those in first webs (x --- 5.8 ± 3.3) were larger than those in second (x = 3.4 ± 3.2) and third (x = 3.0 + 2.7) webs (both p < 0.001 with Mann-Whitney la- test), and the proportion of webs with differences of >4 loops was greater among first than in second or third webs (both p < 0.001 with Chi Square). Since spiders often reused frame lines from previous webs, the sticky spirals of later orbs may have been more symmetrical because these webs tended to have relatively larger frame areas as compared to sticky spiral areas, making fewer turnbacks in the sticky spiral necessary (Eberhard 1969). Experimental modification of relative amounts in glands. — The relative amounts of sticky and non-sticky silk available to the spider when the second web was begun were modified experimentally by allowing the spider to finish adding non-sticky lines to the first web (radii, frame, hub, temporary spiral), but then cutting the radii as above, thus preventing the spider from laying any sticky silk. The spider’s non-sticky silk supply was thereby reduced, while the sticky silk supply was left intact. The experimental second webs that followed did not have reduced areas; they had fewer radii and, in L. mariana , fewer sticky loops than first webs, but the reductions in both were significantly less than those in normal second webs of both species (Table 1). The relative portion of the web enclosed within the outermost sticky spiral loop was not reduced as in normal second webs THE JOURNAL OF ARACHNOLOGY 298 30 DISTANCE OUTER LOOP TO FRAME Fig. 1. — Frequencies of distances between outer loops of sticky spiral and ends of radii in normal first (white), second (hatched) and third (black) webs ( N = 44 in each case) that were spue by L. mariana females (three webs /female, all on the same day). Distributions are significantly different with Chi-Square Test. (predicted distances from outer loops were calculated for experimental webs using the percentages for normal second webs in Fig. 1). By cutting radii as soon as they were laid, L. mariana spiders were experimentally induced to lay more than, one additional set of radii during construction of replacement webs (radii represent approximately 20-30% of the non- -sticky silk in a finished orb — Eberhard 1986), confirming that orb construction of normal second webs does not completely empty the am pul late glands of this species, just as in the first webs of Araneus diadematus Clerck (Kbnig 1951; Witt et aL 1968), L. mariana and Gasteracantha cancriformis (Linneaus) (Eberhard 1986), and the unidentified araneid studied by Kingston (1920). Correlations with spider size and weight. — Correlations between spider size and weight and dimensions of first webs of 138 L. mariana showed that larger and heavier spiders made webs with more closely spaced sticky spiral loops, and somewhat larger webs (Table 2). Partial correlations showed that both weight and a combination of body size measures (x = tibia / average tibia + cephalothorax / average cephalothorax) had significant correlations with sticky spiral spacing (partial correlation coefficients were “”0.26 and —0.23 respectively, both p < 0.001). Partial correlations with web area were not significant. Website effects. — There were strong positive correlations between the slants and areas of successive webs at the same site spue by different spiders (r = 0.46, p < EBERHARD — FLEXIBILITY IN ORB DESIGN 299 Table 2. — Correlation coefficients between web and body measurements of 138 adult female Leucauge mariana (* p < 0.05, ** p < 0.001). Area Number Sticky Loops Space between Sticky Loops Number Radii Slant Cephalothorax 0.206* -0.065 -0.358** -0.056 -0.124 Tibia 0.172 -0.100 -0.343** -0.027 -0.055 Wet weight 0.267** 0.033 -0.399** 0.006 -0.117 0.001 and r — 0.335, p < 0.01 respectively for 63 pairs of webs). Other web features showed no significant relationships, nor was there a significant correlation between the weights of successive spiders occupying the same sites. DISCUSSION It might be thought that differences between first and second webs were due to the first webs usually being spun in darkness and the second in daylight (Ramousse and LeGuelte 1979 on Araneus diadematus). The differences between second and third webs of L. mariana and experimental and second webs of both species (all built during the day) indicate however that light conditions during construction do not explain the changes in design. The reduced numbers of radii in both normal and experimental second orbs of L. mariana and M. sexspinosa suggest that, as in A. diadematus , non-sticky silk in the ampullate glands probably influences their web designs. The experimental webs show, however, that the supply of sticky silk can partially “over-rule” the effect of non-sticky silk availability. A. diadematus may also respond to cues from its sticky silk glands. Preliminary evidence shows similar reductions may occur in radius and sticky spiral loop number and web area when webs are built in close temporal succession (Ramousse 1977). Slight reductions of numbers of sticky spiral loops occurred in webs spun by spiders which had been milked of non-sticky silk, but prevented from laying sticky silk for three weeks (presumably gland output was reduced when demand ceased) as compared with controls which had spun normal webs (21.0 vs. 26.5, N = 6 for both, significance levels were not given) (Reed et al. 1970). Sticky silk forms a large fraction of an orb. Its length ranged from about 36 to 54% of the total length of silk in more or less typical orb designs (Eberhard 1986). By weight it may be an even larger fraction. The non-sticky scaffolds of Argiope aurantia webs weighed only about 16% of the dry weight of the finished orb (Tillinghast et al. 1984). Taking into account the non-sticky stabilimentum, which was included along with the sticky spiral in the remaining 84%, it is probable that the sticky lines account for 70-80% of the dry weight of an orb (E. K. Tillinghast, pers. comm.). Sticky silk is also, of course, a key web component in trapping prey. In sum, it is perhaps not surprising that the supply of sticky silk influences web design. The relatively smaller portions of later webs occupied by sticky spiral, the big proportion occupied in experimental second webs, and the spiders’ ability to lay many extra radii during second web construction all suggest that sticky rather than non-sticky silk sometimes limits web size (see also Eberhard 1986). 300 THE JOURNAL OF ARACHNOLOGY It is tempting to postulate that non-sticky gland contents determine the design of non-sticky web components, and sticky silk supplies affect sticky silk design features. There are, however, probably “crossovers” in cues from the two types of glands; for instance, web area (a design feature of non-sticky lines) was not reduced in experimental second webs of either species, even though supplies of non-sticky silk had been reduced. Individual araneids sometimes produce non-sticky lines with varying diameters (e.g., Christiansen et al. 1962; R. W. Work pers. comm.), and the sizes and spaces between sticky balls on the sticky spiral of L. mariana webs varies substantially even within a single web (Eberhard unpub.). Thus the web measurements given here may not accurately reflect total amounts of material in different webs. The probable trend in diameter modification (smaller diameters when gland less full) suggests that the trends documented here give underestimates of the differences in the amount of material in successive webs. Several previous studies have analyzed the relationships between spider size and weight, but many comparisons include the possibly confounding effects of species or age (instar) differences. Comparisons of webs of conspecific mature females with different weights (Christiansen et al. 1962; Risch 1977) and different body sizes (Risch 1977) suggest that both factors have effects on web design in other araneid species. Variations in L. mariana web designs associated with greater spider weight were similar to those associated with relatively greater supplies of sticky silk (decreased spaces between sticky loops, greater area), but differed in showing no relations with numbers of radii or sticky loops (Table 1). Larger spider weights might be associated to some extent with greater recent feeding success and thus, presumably, greater amounts of material in the glands, but a female spider’s weight is probably largely determined by the stage of development of its eggs. Thus the weight effects documented here may be largely independent of the gland-filling effects. Website may be still another factor causing L. mariana web designs to vary. Spiders often reuse some of the frame lines from previous webs (Eberhard, unpub.), and the correlations in slant and area of successive first webs could be due to frame reuse. It is also possible that other website characteristics were important. Adjustments of web design to local conditions are undoubtedly advantageous for orb weavers; they are suggested by field data (Leborgne and Pasquet 1987), and have been documented in confinement (Tilquin 1942; Szlep 1958; LeGuelte 1966). The intraspecific variations documented here are substantial. For instance, even when one controls for possible effects of spider size, weight, and website in L. mariana , first webs average 146% more radii, 185% more sticky spiral loops, and 166% more area than third webs; some individual spiders, of course, showed even greater variations. The magnitude of this variation, the existence of similar variation in other species (Ramousse and LeGuelte 1979; Leborgne and Pasquet 1987), and the correlations between these and other web characters (Leborgne and Pasquet 1987) weaken the old hope that orb designs can provide reliable species-specific characters (e.g., Savory 1952; Foelix 1982). If such web characters exist, they may be associated with more subtle details such as number and pattern of hub loops, relative size of free zone, etc. (see Coddington 1986 for examples of useful generic characters in webs of Theridiosomatidae). It is possible that different species have different ways of adjusting to changes in factors such as EBERHARD— FLEXIBILITY IN ORB DESIGN 301 supplies of sticky and non-sticky silk and spider size and weight, but proof of this will require much more information that is presently available. Intraspecific variation seen in other studies- (LeGuelte 1966; Risch 1977; Ramousse and LeGuelte 1979, 1984; Nentwig 1983, 1985; Tyschenko and Marusik 1985, Tyshchenko et al. 1985; Buskirk 1986; Leborgne and Pasquet 1987) may be due at least in part to the factors discussed here. In light of these findings, the probable nervous mechanisms controlling orb construction appear to be extraordinarly complex. Both internal factors (weight, body size, contents of sticky and non-sticky silk glands) and external factors (website and / or previous lines present there) are integrated in determining a variety of design features, ranging rom basic characteristics such as numbers of radii and sticky loops and web area, to more subtle aspects such as the relative symmetry of the sticky spiral and the relative fill of the web area with sticky spiral. Different features are modified at least partially independently. The influence of gland contents and perhaps that of the website may incorporate feed- back loops involving amounts of silk and web designs used previously (Reed et al. 1970; Tillinghast and Townley 1986; this study). During actual construction several other factors, such as gravity, leg length and distances and angles between lines, and memories of distances and directions travelled (Kingston 1920; Tilquin 1942; LeGuelte 1966; Vollrath 1986, 1987; Eberhard 1987, 1988) also influence the paths taken and the lines laid. The reasons why spiders opt for different orb designs when they have different amounts of silk available are not yet clear; but there is no doubt that we must discard once and for all the old image of orb weaving spiders spinning out the same rigidly programmed, inflexible geometric patterns in their webs day after day. ACKNOWLEDGMENTS I thank H. W. Levi for identifying the spiders, M. Spivak for measurements of photographs of L. mariana webs, J. A. Coddington, F. G. Stiles, E. K. Tillinghast, M. A. Townley, F. Vollrath, and M. J. West-Eberhard for criticisms of earlier drafts, and the Vicerrectoria de Investigacion of the Universidad de Costa Rica for financial support. LITERATURE CITED Buskirk, R. 1986. Orb-weaving spiders in aggregations modify individual web structure. J. Arachnol., 14:259-265. Coddington, J. A. 1986. The genera of the spider family Theridiosomatidae. Smithsonian Contrib. Zool, 422:1-96. Christiansen, A., R. Baum and P. N. Witt. 1962. Changes in spider webs brought about by mescaline, psilocybin, and an increase in body weight. J. Pharmac. exp. Ther., 136:31-37. Eberhard, W. G. 1961. Computer simulation of orb-web construction. American Zool., 9:229-238. Eberhard, W. G. 1986. Effects of orb-web geometry on prey interception and retention. Pp. 70-100, In Spiders — Webs, Behavior and Evolution (W. A. Shear, ed„). Stanford Univ. Press, Stanford. Eberhard, W. G. 1987. The effect of gravity on temporary spiral construction by the spider Leucauge mariana (Araneae, Araneidae). J. Ethol., 5:29-36. Eberhard, W. G. 1988. Memory of distances and directions moved as cues during temporary spiral construction in the spider Leucauge mariana (Araneae: Araneidae). J. Insect Behav., 1:51-66. Foelix, R. 1982. Biology of Spiders. Harvard Univ. Press, Cambridge. 302 THE JOURNAL OF ARACHNOLOGY Hingston, R. W. G. 1920. A Naturalist in Himalaya. H. F. & G. Willoby, London. Kavanagh, E. and E. K. Tiliinghast. 1979. Fibrous and adhesive components of the orb webs of Araneus trifolium and Argiope trifasciata. J. Morph. , 160:17-25. Konig, M. 1951. Beitrage zur Kenntnis des Netzbaues orbiteiler spinnen, Z. Tierpsychol., 8:462-492. Kovoor, J. 1977. La sole et les giandes sericigenes des arachnides. Ann. Biol, 16:97-171. Leborgne, R. and A. Pasquet. 1987. Influences of aggregative behaviour on space occupation in the spider Zygielia x-notata (Clerck). Behav. Ecol. SociobioL, 20:203-208. LeGuelte, L. 1966. Structure de toile de Zygielia x-notata Cl. (Argiopidae) et facteurs que regissent le comportement de l’araignee pendant la construction de la toile. These Publ. Univ. Nancy, Nancy. Nentwig, W. 1983. The non-filter function of orb webs in spiders. Oecologia (Berlin), 58:418-420. Nentwig, W. 1985. Prey analysis of four species of tropical orbweaving spiders (Araneae: Araneidae) and a comparison with araneids of the temperate zone. Oecologia (Berlin), 66:580-595. Ramousse, R. 1977. Organisation spatio-temporelle du comportement constructed chez Araneus diadematus (Araigneae, Argiopidae). These Univ. Ciaude-Bemard de Lyon, Lyon. Ramousse, R. and L. LeGuelte. 1979. Relations spatio-temporelles dans le comportement constructeur ches YEpeira diademe. Rev. ArachnoL, 2:183-192. Ramousse, R. and L. LeGuelte. 1984. Strategies de construction de la toile chez deux especes d’araignees (Araneus diadematus et Zygielia x-notata). Rev. ArachnoL, 5:255-265. Reed, C. F. and R. L. Jones. 1965. The measuring function of the first legs of Araneus diadematus CL Behaviour, 25:98-119. Reed, C. E, P. N. Witt, M. Scarboro and D. B. Peak a 11. 1970. Experience and the orb web. Develop. PsychobioL, 3:251-265. Risch, P. 1977. Quantitative analysis of orb-web patterns in four species of spiders. Behav. Genet., 7:199-238. Savory, T. H. 1952. The Spider’s Web. Frederick Warne, London. Szlep, R. 1958. Influence of external factors on some structural properties of the garden spider (Aranea diademata ) web. Folia Biol. (Praha), 6:287-299. Tiliinghast, E. K., S. F. Chase and M. A. Townley. 1984. Water extraction by the major ampuliate duct during silk formation in the spider Argiope aurantia Lucas. J. Insect Physiol., 30:591-596. Tiliinghast, E. K. and M. A. Townley. 1986. The independent regulation of protein synthesis in the major ampuliate glands of Araneus cavaticus (Keyserling). J. Insect Physiol., 32:117-123. Tilquin, A. 1942. La Toile Geometrique des Araignees. Presses Univ., Paris. Tyshchenko, Y. P. 1984. The catching webs of orb-weaving spiders 1. The substantiation of the method of standard webs with reference to two species of the genus Araneus (Aranei, Araneidae). (in Russian) Zool. Zhurnal, 63:839-847. Tyshchenko, V. P. and Y. M. Marusik. 1985. Catching webs of orbweaving spiders 3. Geographic variation of webs in Araneus marmoreus (Aranei, Araneidae). (in Russian) Zool. Zhurnal, 64: 18 lb- 1822. Tyshchenko, V. P, , Y. M. Marusik and C. K. Tarabaev. 1985. The catching webs of orb-weaving spiders 2. Comparative study of the webs in the genus Nuctenea (Aranei, Araneidae). (in Russian) Zool. Zhurnal, 64:827-834. Vollrath, F. 1986. Gravity as an orientation guide during web-construction in the orb spider Araneus diadematus. J. Comp. Physiol., 159:275-280. Vollrath, F. 1987. Altered geometry of webs in spiders with regenerated legs. Nature, 328:247-248. Witt, P. N. and R. Baum. 1960. Changes in orb webs of spiders during growth (Araneus diadematus Clerck and Neoscona vertebrata McCook). Behaviour, 16:309-318. Witt, P. N., C. F. Reed and D. B. Peakalh 1968. A spider’s web. Springer Verlag, New York. Manuscript received October 1987, revised January 1988. This paper was included in the Spider Silk Symposium held during the 1987 American Arachnoiogical Society Meeting at Harvard University / a part of the funds raised at this meeting was used to defray publication costs. Townley, M. A. and E. K. Tillinghast. 1988. Orb web recycling in Araneus cavaticus (Araneae, Araneidae) with an emphasis on the adhesive spiral component, GABarnide. J. Arachnol., 16:303- 319. ORB WEB RECYCLING IN ARANEUS CAVATICUS (ARANEAE, ARANEIDAE) WITH AN EMPHASIS ON THE ADHESIVE SPIRAL COMPONENT, GABAMIDE Mark A. Townley and Edward K. Tillinghast Department of Zoology University of New Hampshire Durham, New Hampshire 03824 USA ABSTRACT The feeding of radiolabeled conspecific orb webs to Araneus cavaticus Keyserling clearly demonstrated the ability of this species to solubilize nearly all of the orb web, although in no instance was complete solubilization achieved. A principal component of the nonsolubilized portion is probably minor ampuilate silk, as spiders fed pulled minor ampullate silk were unable to solubilize the majority of the samples. In contrast, spiders were able to completely solubilize pulled major ampullate silk. Despite the overall high percentage of web solubilization, the recycling efficiencies obtained, while variable, were never in excess of 32% (as determined using webs built by spiders fed 14C-glucose). A complete assessment of web recycling will have to consider the fate of ingested low molecular weight adhesive spiral components as well as web proteins. Of those components which have been identified, only GABarnide was followed in the present study due to the labeled compound fed. On average ingested GABarnide appears to be more quickly reincorporated into new web than ingested protein residues and this reutilization is, for the most part at least, in the form of GABarnide. From spiders which did not build webs until several days after being fed orb webs, the indication is that GABarnide can be stored for future web construction for such a length of time. Whether any storage is by physical separation from agents with metabolic activity against GABarnide or by a degree of metabolic inertness is unknown. INTRODUCTION In previous studies the digestive fluid of Argiope aurantia Lucas was found to be capable of solubilizing all orb web adhesive spiral, radiai, and junctional components except minor ampullate fibers (Tillinghast and Kavanagh 1977; Kavanagh and Tillinghast 1979), which commonly accompany major ampullate fibers in radii (Kavanagh and Tillinghast 1979; Work 1981). However, the method employed to make this determination, that of applying filter discs wetted with digestive fluid to plated webs, produced results which, upon reflection, could have been open to misinterpretation. The act of removing the filter disc after the incubation period could have resulted in the simultaneous removal of underlying web components which may have been only partially solubilized or otherwise weakened by the digestive fluid, rather than completely solubilized. Additionally, in no instance was web completely solubilized when immersed in a buffered solution containing digestive fluid (Tillinghast and Kavanagh 1977), and it seemed unlikely that minor ampullate fibers alone could account for all of the nonsolubilized portion. This incomplete solubilization also made the extremely 304 THE JOURNAL OF ARACHNOLOGY high recycling efficiency reported by Peakall (1971) seem unlikely. In an effort to resolve these inconsistencies we have examined orb web digestion and recycling in vivo. The ability to procure individually samples of major ampullate and minor ampullate silk also allowed us to examine more specifically the digestion and, for the former silk type, recycling of these orb web components. It should be noted that while a fraction of the ingested web components was undoubtedly used for purposes not directly related to web construction, but was nevertheless recycled in a denotative sense, in this paper the term “recycle” is restricted to the utilization of those components in subsequent webs or silks. MATERIALS AND METHODS Adult and penultimate female and male Araneus cavaticus Keyserling were collected in southern New Hampshire and Maine and kept either in cages (Tillinghast and Kavanagh 1977) or small vials, depending on whether web construction was desired or not. To obtain radioactive major ampullate silk, spiders were fed from 4 to 20 juCi D-[14C(U)] glucose (sp. act. 4.28 mCi/mmol or 329 mCi/mmol, New England Nuclear®; or 263 mCi/mmol, ICN® Radiochemicals). The silk was mechanically drawn one and/or two days after feeding as described previously (Tillinghast et al. 1984), except that a pulling rate of 1.0 cm/s was used. Again, the silking operation was monitored frequently with the aid of an Olympus®, Model X-Tr, stereo dissecting microscope, to insure as much as possible a collection of major ampullate silk free from pyriform, aciniform, and minor ampullate fibers. Since the aggregate and flagelliform glands of males degenerate shortly after adulthood is reached (Sekiguchi 1955a,b), making adhesive spiral and, thus, orb web construction impossible, adult males were only used for this purpose. The silk pulled from each spider was cut into two portions, one roughly three times the size of the other. Each portion was desiccated over CaSCL and NaOH in vacuo and their dry weights were measured on a Perkin-Elmer® AD-2 autobalance. The smaller portions were hydrolyzed in 6N HCI at 1 10°C for 24 h, with the hydrolysates being used to determine specific activity. Radioactivity was measured in a Beckman® Beta-Mate II scintillation counter using Beckman® Ready-Solv EP as the scintillation fluid, and amino acids were quantitatively measured by the method of Moore (1968). The larger portions were assumed to have the same specific activities as their smaller counterparts. Radioactive whole orb webs were obtained from spiders fed 10 /xCi D-[14C(U)] glucose (sp. act. 263 mCi/mmol, ICN®) each. After collecting the webs on 20 juL micropipettes, each was scraped off as a ring with a new razor blade and cut into one large and one small piece. These pieces were treated the same as the pulled major ampullate silk above. Nonradioactive spiders were fed radioactive pulled silk or whole web in one of two ways. Either the spiders were offered the radioactive material while pinioned or it was placed in the spider’s nonradioactive web. For the former method the spider was temporarily anesthetized with CO2 and taped down, allowing close scrutiny of the external digestion process when the above dissecting microscope was used. For the latter method, all but two opposing frame lines were cut following placement of radiolabeled material in the unlabeled web. This was done TOWNLEY & TILLINGHAST— ORB WEB RECYCLING IN ARANEUS 305 both to encourage web recycling and to insure that none of the radioactive material, particularly the pulled silk, would be lost during the spider’s recycling of the web. The greater freedom of movement permitted by this method created a more normal situation. It was more difficult or impossible, however, to observe the movements of the spider’s mouthparts. Often, two or more radioactive samples were fed to a single spider if an individual sample contained a comparatively low total amount of isotope. Spiders were not fed after ingestion of radioisotope but were given water daily. All subsequent webs built during the remainder of the experimental spiders’ lives were collected for analysis. Note that spiders fed labeled whole orb web or major amp u Hate silk are referred to as web- fed and silk-fed spiders, respectively. In addition to major ampullate silk, the ability of A. cavaticus to digest pulled minor ampullate silk was also examined, though to a much lesser extent since, in our experience, minor ampullate fibers cannot be pulled for long periods of time, as major ampullate fibers can. Also, the small amounts of minor ampullate silk obtainable made radiolabeling and partitioning of the silk impractical. Instead, digestion was only evaluated by observation of pinioned spiders under the dissecting microscope and by comparison of dry silk weights before and after feeding. Some of the webs constructed by spiders fed radioactive material were collected intact on 20.3 cm x 25.4 cm glass plates and placed with Kodak™ SB-5 X-ray film as described previously (Kavanagh and Tillinghast 1979). The remainder of the webs were collected on micropipettes, hydrolyzed, and specific activities determined as described above. In addition, two dimensional thin layer chromatography (2D-TLC) was performed on some of the hydrolysates. Typically, 125 jig leucine equivalent amounts were chromatographed, but occasionally the amount of hydrolysate remaining after specific activity determination necessitated the use of a lesser quantity. For 2D-TLC, 20 x 20 cm Merck® precoated cellulose plates, 0.1 mm thickness, were developed using the solvent systems of Schmidt (1974); pyridine: acetone: ammonium hydroxide:water (45:30:5:20, v/v) for the first dimension and 2-propanol:formic acid (88%):water (75:12.5:12.5, v/v) for the second dimension. Development from the sample origin was 16 cm in both dimensions. Autoradiograms were prepared from the TLC plates as for plated webs, following which amines were visualized using a llmM ninhydrin in acetone solution. In August and September the building of orb webs, particularly by gravid female A, cavaticus , becomes less reliable, at least in the laboratory, than earlier in the season. Typically, these spiders instead lay down a plentiful amount of “random” fibers throughout the cage, which are presumably of major ampullate gland and, secondarily, minor ampullate gland origin, predominantly. Certainly, it seems very unlikely that any aggregate gland material is used in these constructions. Accumulations of “random” fibers produced over one or more days, as well as any orbs built by such silk-fed and web-fed spiders, were collected on micropipettes and treated the same as described above for orb webs. RESULTS Following the digestion of web or silk by pinioned spiders, the remnant present between the endites, if any, was removed and used to estimate the percentage of 306 THE JOURNAL OF ARACHNOLOGY Table 1. — Solubilization of orb webs, major ampullate silk, and minor ampullate silk by pinioned A. cavaticus. The data from spiders fed whole orb webs have been separated into two groups to demonstrate the disparity between them (particularly with respect to the percentages determined by measuring radioactivity). Group 1 spiders were fed web on or between June 23 and July 1 1 and between 2130 and 0645 hours. Group 2 spiders were fed web on or between July 18 and August 23 and between 1100 and 1630 hours. Minhydrin positive compounds were assayed using leucine as the standard. Material Fed to Spiders Percentage of Web or Silk Remaining After Feeding (Mean +SE; Median) n Gravimetric ally Determined As Determined by Measuring Ninhydrin Positive Compounds in Hydrolysates As Determined by Measuring Radioactivity in Hydrolysates Orb Web Group 1 1.6 ± 0.5; 1.6 2.9 ± 1.0; 2.1 0.18 + 0.08; 0.089 5 Orb Web Group 2 3.8 ± 0.6; 3.9 2.9 ± 0.7; 2.6 2.9 ± 0.7; 2.9 4 Maj Amp Silk Only 1.0 ± 0.4; 0.39 0.61 ± 0.22; 0.35 0.12 + 0.10; 0.00 13 Min Amp Silk Only 97 ± 13; 97 — — 2 material which was not solubilized (Table 1). Of the three types of measurement used to make this estimate, that of measuring residual radioactivity was probably the most accurate as it would not have been influenced by non-sample materials incorporated into the remnants; principally hairs loosened from the end lies' scopulae. The higher percentages most often obtained with the other two methods support this belief. Consistent with earlier in vitro studies (Kavanagh and Tilling hast 1979) minor ampullate silk was found to be relatively resistant compared to major ampullate silk. Spiders fed minor ampullate silk were unable to solubilize a large majority of the samples despite digestion attempts which were typical of spiders fed major ampullate silk or whole web, both in terms of method used and time involved. The contribution of non-sample inclusions to the remnants was made apparent in one of the two minor ampullate silk feedings by the weight of the remnant exceeding that of the original sample. In the second feeding 16% of the sample, by weight, was solubilized. Again, this figure may be somewhat low due to non-sample contaminants. Additionally, the former silk sample was subsequently fed to six successive spiders, wrapped each time with a new orb web to further encourage digestion. The remnant left by the sixth spider had a weight approximately three times that of the original sample, with contributions from the orb webs no doubt accounting for much of the increase in weight. Nevertheless, the original sample, which was whiter than the rest of the remnant, could still be distinguished and was not noticeably diminished by the six spiders. By contrast, in eight out of thirteen major ampullate silk feedings either no remnant was left at all or no significant radioactivity (< 3 SD from the mean background count) was detectable in the remnant. In all cases a very large fraction of the sample was solubilized. Unexpectedly, radioactivity measurements made on remnants from orb web feedings indicated two significantly different groups (approximate cJ.es L Sokal and Rohlf 1981; P < 0.05; Table 1). An examination of the feeding conditions for each of the spiders in these two groups revealed two consistent differences; the time of day and year during which feeding took place. Group 1 spiders were fed web on or between June 23 and July 11 and between 2130 and 0645 hours. TOWNLEY & TILLINGHAST— ORB WEB RECYCLING IN ARANEUS 307 Table 2. — Efficiency in recycling ingested l4C-labeled web and major ampullate silk components. Range and mean values reflect the total isotope present in all webs and/or “random” fibers collected from a given spider subsequent to feeding, expressed as a percentage of the total amount of isotope present in the material fed to the spider. Radioactive Material Fed to Spiders Material Collected and Analyzed 14C-Radioactivity Recycled Range (%) Mean (%) SE (%) n Orb Web Orb Webs 4.00-32.0 16.3 2.93 12 Orb Web Random Fibers, Orb Webs 4.20-6.09 5.1 0.95 2 Major Ampullate Silk Orb Webs 0.430-23.2 10.8 4.86 4 Major Ampullate Silk Random Fibers, Orb Webs 2.91-23.7 13.2 1.85 12 Total 0.430-32.0 13.6 1.57 30 Group 2 spiders were fed web on or between July 18 and August 23 and between 1100 and 1630 hours. Whether either or both of these differences were involved in producing the different solubilization percentages is unknown. Features common to both groups include a high percentage of solubilization, although, on average, not as high as for major ampullate silk, and incomplete solubilization in all cases (radioactivity in remnants > 5 SD from the mean background count). The efficiency with which A. cavaticus was found to recycle ingested whole web or major ampullate silk is presented in Table 2. Total isotope present in all webs and/or “random” fibers produced by each spider after being fed radioactive material are expressed as a percentage of the total isotope present in the radioactive material fed. Considerable variability in this percentage was apparent, irrespective of the method used to feed the spiders. In no instance, however, did our recycling percentages even approach those determined by Peakall (1971) for Araneus diadematus Cl. Whereas we obtained a maximum recycling of 32%, which takes into account the total amount of isotope present in all webs constructed by the spider, Peakall (1971) typically found the percentage of recycled material in the first web constructed to be in excess of 90%. Earlier estimates of recycling for A. diadematus made by Breed et al. (1964) were more in keeping with our results, ranging from 21 to 50%. Their estimates were made from the total radioactivity present in the first two webs constructed by each spider. The normalized specific activities of webs built by twelve spiders fed radioactive whole web and by four spiders fed radioactive major ampullate silk are presented in Figs. 1 and 2, respectively. Actual peak specific activities ranged from 2.0 to 47 CPM//ug leucine equivalents in Fig. 1 and from 1.2 to 79 CPM//xg leucine equivalents in Fig. 2. Taking into consideration the amount of isotope contained in the material fed, peak specific activities ranged from 9.8 X 10"6 to 3.3 X 10“ CPM/(CPM in web fed X qg Leu equiv.) in Fig. 1 and from 1.9 X 10~5 to 2.7 X 10”4 CPM/(CPM in silk fed X /jig Leu equiv.) in Fig. 2. For eleven of the twelve web-fed spiders, peak specific activity was present in the first web constructed, while all four silk-fed spiders attained peak specific activity in the second web built. 2D-TLC and subsequent autoradiography of hydrolysates prepared from these webs would indicate that GABamide (4-aminobutyramide) played a major role in producing this difference. GABamide is present in the adhesive spirals of those Araneidae examined thus far (Fischer and Brander 1960; Anderson and Tillinghast 1980; Tillinghast and Christenson 1984). Particularly strong support 308 THE JOURNAL OF ARACHNOLOGY Time After Feeding Whole Web (Days) Fig. 1. — Incorporation of isotope into webs built by twelve spiders fed 14C-labeled whole orb webs. The surge in specific activity at web 7 (day 28) from the spider represented by solid triangles was due entirely to isotope incorporation into UC8, as revealed by autoradiography of the 2D-TLC plate prepared from this web’s hydrolysate. Fig. 2. — Incorporation of isotope into webs built by four spiders fed 14C-labeled major ampullate silk. TOWNLEY & TILLINGHAST— ORB WEB RECYCLING IN ARANEUS 309 c © m c m E b w Origin 2nd Dimension — — — ► Fig. 3. — 2D-TLC of three pooled web 1 hydrolysates from web-fed spiders. Of the 125 jug leucine equivalents chromatographed, 34 jug were from a web with a sp. act. of 47 CPM/jug Leu equiv. (built 1 day after feeding, represented in Fig. 1 by a solid diamond), 77 jug were from a web with a sp. act. of 30 CPM/jug Leu equiv. (built 1 day after feeding, represented in Fig. 1 by an open square), and the remaining 14 jug were from a web with a sp. act. of 43 CPM/jug Leu equiv. (built 2 days after feeding, represented in Fig. 1 by an open diamond). An exposure of 110 days was used to produce the autoradiogram (right). A = alanine; D = aspartic acid; E = glutamic acid; G = glycine; GABA = 4- aminobutyric acid; I = isoleucine; L = leucine; P = proline; T = threonine; V = valine. Proline has been circled in the chromatogram since a yellow product, difficult to see in black and white photographs, is formed when proline is reacted with ninhydrin. for this proposal came from the first webs built by three of the web-fed spiders, two of which were built 1 day after feeding and one which was built 2 days after feeding. In these webs’ hydrolysates virtually all of the isotope was restricted to GABA (4-aminobutyric acid; Fig. 3), the hydrolytic product of GABamide. Less extreme results were obtained from the other hydrolysates chromatographed (Figs. 4, 5, 6). For the spider whose webs are presented in Fig. 4, GABamide was still the major radioactive compound present in the first web built after feeding, but some amino acids and as yet unidentified compounds were also carrying label. Note that web 1 was built 5 days after feeding. In Fig. 5, GABA and an unidentified compound (UC1) can be seen to possess comparable amounts of isotope in web 1, with other unidentified compounds containing considerably lesser amounts. Web 1 of Fig. 6, built 6 days after feeding, again shows GABA dominating the autoradiogram. Thus, despite the concurrent high specific activity of UC1 in one instance, it was GABamide which appeared to be responsible for the maximum specific activities most often occurring at web 1 in web-fed spiders. The single exceptional web-fed spider produced atypical webs having few or no adhesive spiral loops. In the chromatogram prepared from this spider’s first web GABA was only barely discernible and in chromatograms prepared from subsequent webs GABA was not visible at all. In contrast, while GABA could clearly be seen in 2D-TLC autoradiograms prepared from the second webs built by silk-fed spiders, it certainly did not carry the majority of the label (Figs. 7, 8). Rather, several of the amino acids prevalent in web proteins were evidently responsible for the peak specific activities in the second webs of these spiders. 310 THE JOURNAL OF ARACHNOLOGY Webl Day5 e o c a> E a Fig. 4. — 2D-TLC of hydrolysates (125 jug Leu equiv.) made from the first two webs built by a web- fed spider. These webs are represented in Fig. 1 by solid triangles. Autoradiograms form the right side of each pair. Sp. act. (CPM/jug Leu equiv.): web 1, 14; web 2, 3.6. X-ray film exposures (days): web 1, 119; web 2, 122. A = alanine; D = aspartic acid; E = glutamic acid; F = phenylalanine; G = glycine; GABA = 4-aminobutyric acid; I = isoleucine; K = lysine; L = leucine; P = proline; S = serine; T = threonine; V = valine. Numbers designate unidentified compounds (UC). Autoradiograms prepared from plated webs were consistent with the TLC results. Thus, the first webs built by three web-fed spiders had adhesive spirals which were much more intensely labeled than radii or hub spirals. The adhesive coverings were particularly dark, especially considering the extent to which they were smeared during preparation for autoradiography. Subsequent webs had less intense adhesive coverings and were apparently less labeled overall. In the second webs built by two silk-fed spiders, the radii and adhesive spirals were of roughly equal intensity and the adhesive spiral core fibers appeared to contain the majority of the adhesive spirals’ isotope. Peak activity was apparently possessed by these second webs. Also instructive were the results obtained when collections of “random” fibers, in addition to any webs constructed, were analyzed. The normalized specific activities of such collections and webs, produced by two web-fed spiders, are presented in Fig. 9. Likewise, those produced by ten silk-fed spiders are shown in TOWNLEY & TILLINGHAST— ORB WEB RECYCLING IN ARANEUS 311 Web 1 Day 1 Fig. 5. — 2D-TLC of hydrolysates (125 /ug Leu equiv.) made from the first two webs built by a web- fed spider. These webs are represented in Fig. 1 by solid squares. An exposure of 1 1 1 days was used to produce the autoradiograms (right side of each pair). Sp. act. (CPM/^tg Leu equiv.): web 1, 38; web 2, 12. See Fig. 4 for explanation of symbols. Fig. 10. Unlike the trend observed in Fig. 1, the maximum specific activity was not present in the first collection of “random” fibers from either of the web-fed spiders (Fig. 9). Assuming the construction of “random” fibers does not involve aggregate gland secretions, it would seem that this difference was due to the lack of an outlet for ingested radioactive GABamide as GABamide. Note that one of the web-fed spiders built nine webs following the collection of the first “random” fibers and that the first web built possessed the highest specific activity. 2D-TLC and autoradiography of this first web revealed that while radioisotope was clearly present in glycine, alanine, glutamic acid, aspartic acid, and serine, GABA and two unidentified compounds (UC2, UC3) contained the majority of the isotope (Fig. 11). GABA’s intensity on the autoradiogram was particularly striking considering the relatively low amount of GABA demonstrated by the chromatogram. Apparently, with the building of the first web, an outlet for GABamide was provided, resulting in peak activity. Actual peak specific activities in Figure 9 were 0.28 CPM/jug leucine equivalents [3.8 X 10"6 CPM/(CPM in 312 THE JOURNAL OF ARACHNOLOGY Origin 1 f GABA 2ncl Dimension — ► Fig. 6. — 2D-TLC of a web 1 hydrolysate (125 jug Leu equiv.) from a web-fed spider. The web, built 6 days after feeding and having a sp. act. of 11 CPM/^ig Leu equiv., is represented in Fig. 1 by an open triangle. An exposure of 119 days was used to produce the autoradiogram (right). See Fig. 4 for explanation of symbols. web fed X /ig Leu equiv.)] for the spider from which only “random” fibers were collected and 12 CPM//xg leucine equivalents [5.2 X 10~5 CPM/(CPM in web fed X iig Leu equiv.)]. Of the ten silk-fed spiders (Fig. 10), peak activity was present in the first collections of “random” fibers produced by three of these spiders, in the second collections produced by six of the spiders, and in the third collection produced by the tenth spider. Recalling that peak activity in second webs built by spiders fed major ampullate silk was apparently due to several common web protein residues (Figs. 7, 8), it is not surprising that peak activity for the majority of the spiders in Fig. 10 occurred in the second or third “random” fiber collections. Such protein residues are obviously utilized in “random” fibers as well as in webs. The occurrence of peak activity in first and third collections may simply reflect the variability inherent in the time between the synthesis of progenitive silk components and their inclusion in drawn silk fibers. Alternatively, while it was our intent to obtain comparable amounts of “random” fibers during each collection, we found that similar appearing fiber accumulations in the cages sometimes possessed deceptively and considerably different weights. Most probably, both of these sources of variability contributed to the observed results. In contrast to the single case in Fig. 9, no silk-fed spider produced a web with a specific activity significantly higher than the “random” fibers produced just prior to it, indicating that ingested radioactive GABamide was the source of most of the radioactive GABamide seen in the first web of Fig. 9. Actual peak activities in Fig. 10 ranged from 0.077 to 15 CPM//ug leucine equivalents [3.2 X 10~6 to 3.4 X 10"5 CPM/(CPM in silk fed X jug Leu equiv.)] DISCUSSION It is clear from the web feeding trials that A. cavaticus is able to solubilize the vast majority of the orb web, a fact which was not revealed by the in vitro studies TOWNLEY & TILLINGHAST— ORB WEB RECYCLING IN ARANEUS 313 Web 1 Day 1 Origin 2n^ Dimension — GABA Web 2 Day 2 Fig. 7. — 2D-TLC of hydrolysates made from the first two webs built by a silk-fed spider. The amounts chromatographed were 125 ;ug Leu equiv. from web 1 and 140 fig Leu equiv. from web 2. These webs are represented In Fig. 2 by open triangles. An exposure of 1 1 1 days was used to produce the autoradiograms (right side of each pair). Sp. act. (CPM/jug Leu equiv.): web 1, 3.1; web 2, 12. See Fig. 4 for explanation of symbols. on rod-wound web (Tiliinghast and Kavanagh 1977). Not only was a greater percentage of the web solubilized in vivo , but at a clearly greater rate, such that more digestion occurred in vivo within 20 min than occurred in vitro within 24 h. Certainly these differences must have been in part a result of the digestive fluid dilution made during the in vitro studies; a factor which may have been important not just because of the lowered protease concentration. As proposed earlier (Kavanagh and Tiliinghast 1983), digestion may also require or be facilitated by non-enzymatic components in the digestive fluid, such as surfactants, which would also have been diluted. In addition, observations on pinioned spiders indicate that the contribution of mastication and digestive fluid replenishing to orb web digestion is considerable. Spiders frequently rotated, pierced, and compressed web or silk samples using their fangs and endites, and, often at very short intervals, ingested the digestive fluid already surrounding a sample, only to regurgitate more digestive fluid immediately thereafter. These 314 THE JOURNAL OF ARACHNOLOGY Web 1 Day 1 Web 3 Day 5 Fig. 8. — 2D-TLC of hydrolysates made from the first three webs built by- a silk-fed spider. The amounts chromatographed were 125 jug Leu equiv. from webs 1 and 3, and 60 ng Leu equiv. from web 2. These webs are represented in Fig. 2 by solid squares. An exposure of 122 days was used to produce the autoradiograms (right side of each pair). Sp. act. (CPM/jug Leu equiv.): web 1, 28; web 2, 79; web 3, 22. See Fig. 4 for explanation of symbols. TOWNLEY & TILLINGHAST— ORB WEB RECYCLING IN ARANEUS 315 Time After Feeding Whole Web (Days) Fig. 9. — Incorporation of isotope into webs and “random” fibers produced by two spiders fed 14C- labeled whole orb webs. R = collections of “random” fibers. actions presumably helped to hasten and maintain exposure of the entire sample to active enzyme. Despite the greater extent of orb web solubilization in vivo , complete solubilization was still never achieved. Unlike the situation in vitro , however, the small percentage of nonsolubilized web which remained in vivo does not preclude the possibility that the remnants were composed primarily of minor ampullate silk. In fact, the percentage of web remaining after some feedings was so small as to indicate that minor ampullate silk must be at least partially digestible. The possibility that these webs may simply have contained very few or no minor ampullate fibers cannot be excluded, but seems unlikely based on observations of minor ampullate fiber occurrence in orb webs (Kavanagh and Tillinghast 1979; Work 1981). Moreover, evidence for partial digestion was also obtained in one of the two minor ampullate silk feedings. Kovoor (1972) has demonstrated the composite nature of minor ampullate fibers from A. diadematus through a comparison of the distal and proximal minor ampullate cell types. The granules secreted into the lumen by these two cell types were found to be histochemically distinct. This raises the possibility that partial digestion could result from the selective digestion of one or more of the component species of minor ampullate fibers. At present we cannot explain the large discrepancy between the recycling efficiencies we obtained and those of Peakall (1971). A number of differences in the materials and methods used could have contributed to this discrepancy. These differences included the species of Araneus used, the radiolabeled compound fed, and the method used to estimate the total amount of isotope in the ingested web. Also, Peakall considered the time between web recycling and new web construction to be critical to efficient recycling; a time which in A. diadematus was reportedly not more than one hour. As a consequence of the methods we Time After Feeding Maj Amp Silk (Days) Fig. 10. — Incorporation of isotope into webs and “random” fibers produced by ten spiders fed l4C- labeled major ampullate silk. All data points in the top graph were obtained from “random” fiber collections. In the center and bottom graphs, an R designates “random” fiber collections. The ten spiders were separated into three graphs merely for the sake of clarity. Note the different ranges of the abscissas. TOWNLEY & TILLINGHAST— ORB WEB RECYCLING IN ARANEUS 317 Origin 2nt^ Dimension — — — — ► Fig. 11. — 2D-TLC of a web 1 hydrolysate (125 /zg Leu equiv.) from a web-fed spider. The web, built 6 days after feeding and after one “random” fiber collection was made, is represented in Fig. 9 by an open square. Sp. act. 12 CPM//zg Leu equiv. An exposure of 117 days was used to produce the autoradiogram (right). See Fig. 4 for explanation of symbols. used to feed web, this interval was either over ten hours or under ten hours but unknown in our experiments on recycling efficiency. Thus, additional experiments in which this interval is shortened will be required to better evaluate Peakall’s claim. Solely on the basis of the digestibility of the orb web as determined in vivo , and contrary to the previous in vitro findings, the high recycling efficiency reported by Peakall is at least plausible. From the specific activities of the successive webs built by spiders fed radioactive web or silk, along with the chromatograms and autoradiograms prepared from those webs’ hydrolysates, it would appear that on average ingested GABamide is reutilized in new web more quickly than ingested protein residues. Thus, the first web constructed by a spider fed whole web usually had a higher specific activity and more total isotope than webs produced subsequently, and the relatively high specific activity of GABamide in the first web was apparently responsible for this trend. In contrast, for spiders fed major ampullate silk, peak activity and the largest total amount of isotope were present in the second webs constructed, and protein residues, particularly alanine, glycine, glutamic acid, and serine, possessed a large majority of the isotope in these webs. The results from web-fed spiders which produced constructions lacking GABamide (i.e., “random” fibers; Fig. 9) were more similar to those from silk-fed spiders and lend further support for GABamide’s more rapid reutilization. As radiolabeled GABamide was present in the webs of spiders fed major ampullate silk (Figs. 7, 8), it is reasonable to assume that non-GABamide web components were also used to synthesize some of the labeled GABamide present in webs built by web-fed spiders. However, since GABamide was responsible for peak specific activity only in webs built by web-fed spiders, it is also reasonable to assume that most of the radioactive GABamide in these webs must have come from radioactive GABamide in the ingested web. The results also indicate that a sizable fraction of the ingested GABamide may remain available for incorporation into new web for at least several days, should web construction be forgone for such a period. Whether this is due to an actual 318 THE JOURNAL OF ARACHNOLOGY sequestration of GABamide or a resistance to metabolic conversion or both cannot be stated. Whatever the cause, it was found that the first web built by a web-fed spider could still, as a result of GABamide, have peak specific activity and the largest total amount of isotope even if 5 (Fig. 4) or 6 (Fig. 6) days elapsed between feeding and its construction. Somewhat similar results were obtained from another web-fed spider despite a substantial quantity (1.16 mg Leu equiv.) of “random” fibers being laid down before web l’s construction; which was 6 days after feeding (Fig. 11). However, in this instance GABamide, UC2, and UC3 were each influential in producing the maximum specific activity. Again, that the majority of GABamide’s label in these three webs was from ingested GABamide is indicated by the results from silk-fed spiders; in particular, the observation that isotope in the web with the highest specific activity was not localized primarily in GABamide. Due to the radiolabeled compound used, no data were obtained on other known components of the adhesive spiral’s covering, such as the inorganics, KH2PO4 and KNO3 (Schildknecht et al. 1972). This was also true for taurine, obtained by acid hydrolysis from the adhesive spiral’s taurine derivative(s) (Fischer and Brander 1960; Anderson and Tillinghast 1980), since 14C-labeling of this compound was meager at best in spiders fed radiolabeled web or silk. Thus, it is not known if GABamide’s behavior is shared by other low molecular weight adhesive spiral components. During the course of the 2D-TLC, several unidentified compounds have repeatedly been encountered and designated UC1-UC10 (Figs. 3-8, 11). UC1-UC4, UC7, UC8, and UC10 are ninhydrin negative but can incorporate isotope when spiders are fed 14C-glucose. UC5 and UC6 are ninhydrin positive but have not been found to incorporate isotope. UC9 is neither ninhydrin positive nor does it become radioisotopically labeled by 14C=glucose. However, the cellulose support of the TLC plates fortuitously takes on a light purple background hue with ninhydrin visualization, which UC9 inhibits. Thus, within about 1 day after application of the ninhydrin spray, UC9 makes its presence known by the white spot it leaves on the chromatogram. UC3 behaves similarly to UC9 in this respect. ACKNOWLEDGMENTS This work was supported in part by funds from the American Heart Association (N.H. affiliate), as well as CURF and BRSG funds from the University of New Hampshire. LITERATURE CITED Anderson, C. M. and E. K. Tillinghast. 1980. GABA and taurine derivatives on the adhesive spiral of the orb web of Argiope spiders, and their possible behavioural significance. Physiol. Ent., 5:101- 106. Breed, A. L., V. D. Levine, D. B. Peakall and P. N. Witt. 1964. The fate of the intact orb web of the spider Araneus diadematus Cl. Behaviour, 23:43-60. Fischer, F. G. and J. Brander. 1960. Fine Analyse der Gespinste der Kreuzspinne. Hoppe-Seyler’s Z. Physiol. Chem., 320:92-102. TOWNLEY & TILLINGHAST— ORB WEB RECYCLING IN ARANEUS 319 Kavanagh, E. J. and E. K. Tillinghast. 1979. Fibrous and adhesive components of the orb webs of Araneus trifolium and Argiope trifasciata. J. Morph., 160:17-32. Kavanagh, E. J. and E. K. Tillinghast. 1983. The alkaline proteases of Argiope. II. Fractionation of protease activity and isolation of a silk fibroin digesting protease. Comp. Biochem. Physiol., 746:365-372. Kovoor, J. 1972. Etude histochimique et cytologique des glandes sericigenes de quelques Argiopidae. Ann. Sci. Nat. Zool. Biol. Anim., 14:1-40. Moore, S. 1968. Amino acid analysis: Aqueous dimethyl sulfoxide as solvent for the ninhydrin reaction. J. Biol. Chem., 243:6281-6283. Peakall, D. B. 1971. Conservation of web proteins in the spider, Araneus diadematus. J. Exp. Zool., 176:257-264. Schildknecht, H., P. Kunzelmann, D. Krauss, and C. Kuhn. 1972. fiber die Chemie der Spinnwebe, I. Arthropodenabwehrstoffe, LVII. Naturwissenschaften, 59:98-99. Schmidt, L. 1974. The biochemical detection of metabolic disease: Screening tests and a systematic approach to screening. Pp. 675-697, In Heritable Disorders of Amino Acid Metabolism: Patterns of Clinical Expression and Genetic Variation, (W. L. Nyhan, ed.). Wiley, New York. Sekiguchi, K. 1955a. Differences in the spinning organs between male and female adult spiders. Sci. Rep. Tokyo Kyoiku Daigaku Sect. B, 8:23-32. Sekiguchi, K. 1955b. The spinning organs in sub-adult geometric spiders and their changes accompanying the last moulting. Sci. Rep. Tokyo Kyoiku Daigaku Sect. B, 8:33-40. Sokal, R. R. and F. J. Rohlf. 1981. Biometry. Freeman, New York, 859 pp. Tillinghast, E. K. and T. Christenson. 1984. Observations on the chemical composition of the web of Nephila clavipes (Araneae, Araneidae). J. Arachnol., 12:69-74. Tillinghast, E. K. and E. J. Kavanagh. 1977. The alkaline proteases of Argiope and their possible role in web digestion. J. Exp. Zool., 202:213-222. Tillinghast, E. K., S. F. Chase and M. A. Townley. 1984. Water extraction by the major ampullate duct during silk formation in the spider, Argiope aurantia Lucas. J. Insect Physiol., 30:591-596. Work, R. W. 1981. Web components associated with the major ampullate silk fibers of orb-web- building spiders. Trans. American Microsc. Soc., 100:1-20. Manuscript received January 1988, revised March 1988. This paper was included in the Spider Silk Symposium held during the 1987 American Arachnological Society Meeting at Harvard University; a part of the funds raised at this meeting was used to defray publication costs. Starr, C. K. 1988. Sexual behavior in Dictyna volucripes (Araneae, Dictynidae). J. Arachnol., 16:321- 330. SEXUAL BEHAVIOR IN DICTYNA VOLUCRIPES (ARANEAE, DICTYNIDAE)1 Christopher K. Starr2 Department of Entomology University of Kansas Lawrence, Kansas 66045 USA ABSTRACT Courtship and mating in Dictyna volucripes Keys, are described on the basis of laboratory observations of 13 virgin pairs. Their behavior conformed well to the general pattern within the family. Various features of both female and male behavior are consistent with the view that courtship functions mainly in influencing mate-choice by females, rather than in inhibiting predatory attack upon males. Laboratory and field observations show that pairs commonly remain together for some days after mating. While the function of such cohabitation is unknown, it can evidently provide an important preadaptation in the evolution of spider sociality. INTRODUCTION The Dictynidae is a widespread family of small to medium-sized, cribellate spiders which make irregular webs. In recent years much attention has focused on the permanently social Mallos gregalis (Simon), which has in turn called comparative attention to the behavior of more typically solitary or intermediate species (Honjo 1977; Jackson 1977-1979; Uetz 1983). Observations of courtship and mating have been reported from about 12 species of Dictynidae (Karpinski 1882; Montgomery 1903; Berland 1916; Gerhardt 1924; Locket 1926; Billaudelle 1957; Leech 1966; Bristowe 1971; Jackson 1979). Before Jackson’s (1979) analysis of sexual behavior in two Mallos species and Dictyna calcarata Banks, observation was mostly rather superficial, with few quantitative data. As a result, comparisons based on the older literature are often inconclusive. Jackson (1979) has reviewed sexual behavior in the family. In this paper I describe courtship and mating in an additional species, with some remarks on post-mating cohabitation. Dictyna volucripes Keys, is widespread in eastern North America (Chamberlin and Gertsch 1958) and often locally abundant. In eastern Kansas I have found the web typically in the upper part of a small plant, where it forms an irregular tent over a flowerhead or several twigs. A small region of the interior is ’Contribution no. 1974 from the Department of Entomology, University of Kansas, Lawrence, Kansas 66045. This paper is dedicated to Willis J. Gertsch on the occasion of his 80th birthday, 4 October 1986. Dr. Gertsch has been very generous to me and many other amateur arachnoiogists. 2Present address: Department of Horticulture, University of Georgia, Athens, Georgia 30602 USA. 322 THE JOURNAL OF ARACHNOLOGY Fig. 1. — Part of a D. volucripes web with a mating pair in the retreat area. The opening to the retreat is in the middle foreground. reinforced with silk to form a distinct, tubular retreat (Fig. 1). The spider is most often found motionless within the retreat. For a clear illustration of web structure in a related species, see Bristowe (1971: Fig. 41). Preliminary observations in old fields in eastern Kansas indicate that D. volucripes usually overwinters in the subadult stage and that males molt to adulthood a few days before females. Sexual dimorphism is not pronounced, with adult females only slightly larger than males. Scheffer (1905) reported the appearance of egg-sacs in the webs from late June to late September in this area, with 15 eggs/ sac and usually 1-5 sacs/ web. MATERIALS AND METHODS Female and male subadult spiders were collected in northeastern Kansas in March-April of 1975 and 1976. Individuals were reared to adulthood in separate vials, so that all were known to be virgin when first paired. Both as subadults and adults, spiders were provided with flies ( Drosophila sp.) as prey and appeared well fed, except as otherwise noted. Newly emerged females were introduced onto separate dry, tree-like plant stalks (henceforth called “trees”) which simulated wild web-sites and then left to spin webs. Each tree was held upright in sand and covered with a large glass jar, so that it was free on all sides. After 1-3 days, the jar was removed, an adult male introduced at the tree base, and behavior noted with the aid of a tape recorder and hand lens. After a pair had shown no apparent sexual behavior for at least 30 min, we ended observation and replaced the jar. The pair was checked daily for the next four days for a general indication of its condition. STARR — DICTYNA SEXUAL BEHAVIOR 323 Specimens from Kansas collected and determined by C. K. Starr in 1975-1976 can serve as vouchers. These are deposited in the Snow Entomological Museum at the University of Kansas and the Canadian National Collection in Ottawa. RESULTS The following account is based on observation of 13 pairs. In three of these the female had been kept without food for up to two weeks; in all others both partners were well fed. An additional pair which showed no apparent pattern of sexual behavior is disregarded. Courtship. — At the time of male introduction, the female was usually in the retreat in the at-rest posture (Fig. 4): motionless, body lying against the substrate and the legs drawn in close. The male usually began immediately to climb the tree and always reacted strongly upon touching the female’s silk. Typically, he walked extensively on the outside of the web, laying down silk. Such ranging-spinning was usually rapid and often had a notably agitated appearance. The abdomen twitched up and down, and the pace of walking was very uneven. The palps were held in front, alternately lowered and raised. The female’s first reactions to ranging-spinning could in each case be interpreted as alertness to a potential prey or intruder. She came out of at-rest, extending her legs and raising her body off the substrate. Often she walked out of the retreat, and in some cases rushed toward the male, though without coming very close. As ranging-spinning proceeded, the female showed less and less reaction, and in most cases she entered the retreat and returned to at-rest within a very few minutes. After several minutes, ranging-spinning gave way to a new phase, local- spinning, in which the male walked much more closely around the female and sometimes came to walk directly upon her. In two trials the female moved a short distance away from the male, but in others she remained still. Local-spinning evidently added silk to the retreat, as this came to appear denser. When the male walked upon the female, it appeared from movements of his abdomen that he bound her very lightly with silk. During both ranging-spinning and local-spinning, males showed little response to female behavior. I could see no reaction when a female simply became alert inside the retreat or walked just outside it. The few times that a female rushed at the male, he retreated on the web, to remain briefly inactive before resuming ranging-spinning. In no case did the female chase a retreating male. Two trials were performed with a male released at the base of a tree from which the female was newly removed, in order to see his reactions to the web alone. In each case, he went through ranging-spinning and local-spinning as if a female were present. Local-spinning was followed by a phase in which the pair remained in direct physical contact. In all trials this began with the male coming face to face with the female, their faces apparently touching. This was followed by a period, usually lasting a few minutes, in which the male stroked the female’s cephalothorax and parts of her legs with his palps, forelegs, second legs, and occasionally his third legs. As this proceeded, he appeared to attempt to raise her venter away from the substrate with his legs, and in trials which included mating 324 THE JOURNAL OF ARACHNOLOGY Fig. 2. — Mating pair of D. volucripes. the stroking-phase ended with her rising up. In some trials, this phase was interrupted by a brief return to local-spinning, and in some the female broke contact and moved slightly away, in which case the male local-spun for a time before resuming face-to-face contact and stroking. Female behavior in the stroking-phase, where she did not break contact, appeared almost entirely passive. She never stroked the male and at most drew her legs in still closer to the body. Mating. — Raising of the female by the male was always quickly followed by a palpal insertion and was evidently a necessary prelude to it. In three trials without raising there was no insertion, even though in one of these the male courted for more than an hour. In 10 trials with insertion, courtship (comprising the ranging-spinning, local-spinning and stroking phases) lasted for 10-93 min, with a mean of 30 min. In five of nine trials the first insertion was with the left palp, while in four it was the right; in the 10th trial it was not noted. The mating position in all cases was a variant of Gerhardt and Kaestner’s (1937) position I (Figs. 2-3). The male’s face was toward the female’s sternum, so that the two bodies formed an approximately right angle. The male was rotated to one side, so that one palp was closer than the other to her epigynum, and this palp was inserted. The period of first continuous insertion was very variable, lasting 1-109 min (mean = 56 min, SD = 36 min). During mating, the hematodocha of the palp pulsated rhythmically. Most of the time it was dilated, with very brief, strong contractions at intervals. In 15 insertions in which a sample of pulsations was timed, the mean interval between contractions was 6.7 s (range = 2-12 s, SD = 2.2 s), with most samples in the 7-9 s range. One male showed an unusual pattern of pulsations during two insertions: after a series of regular, brief contractions, the hematodocha remained contracted for several seconds before the next series of regular contractions. STARR — DICTYNA SEXUAL BEHAVIOR 325 Further courtship and mating. — At the end of the first insertion phase, the pair disengaged simply and directly. Subsequent behavior was less predictable than that leading up to insertion. The observed variants can be divided into four groups: a. No sexual activity. In two trials the pair soon became motionless and remained so for the rest of the observation period. This was also the usual pattern following final insertion in the next two variants, the pair remaining at rest in or near the retreat (Fig. 4). One of the two pairs was noticed mating again the next day. b. Courtship without mating. In two trials the male resumed local-spinning for a short time, though without subsequent stroking. Courtship in these cases seemed weak and progressively disorganized. c. Courtship with mating. In three trials resumed courtship culminated in insertion of the other palp. In one of these, ranging-spinning preceded local- spinning and stroking. Respective durations of courtship were 12, 25 and 32 min. In the latter the female seemed resistant, as the male made several apparent attempts to raise her before he succeeded. In another of these trials the male again inserted the first palp almost immediately after his second withdrawal, without a return to courtship. d. Mating without courtship. In the remaining three trials the male inserted the second palp without a prior return to courtship. The intervening period was at most about 4 min. In one of these trials the spiders then remained at rest for 79 min, after which the male again inserted the first palp, and almost immediately upon withdrawing it he again inserted the second palp. Another trial was marked by extreme brevity in both insertions and apparent strong unreceptivity of the female after the second withdrawal. It was unclear which partner actively broke contact, but immediately after the first withdrawal the male briefly and unsuccessfully attempted to re-insert the same palp. After the second withdrawal he courted intermittently but vigorously for more than an hour, without further mating that day. The eight recorded second and subsequent insertions had durations of 4-60 min (mean = 31 min, SD - 23 min). If we disregard the (apparently anomalous) one- minute first insertion, these eight later insertions are significantly shorter (/-test, P < 0.05). 326 THE JOURNAL OF ARACHNOLOGY Cohabitation. — We made no systematic observations on the tendency of females and males to occupy webs together, but there are indications that they may commonly do this for extended periods. In the laboratory, pairs left undisturbed after mating remained without apparent conflict during the four days of observation, much of the time together in the retreat. Although the spiders were confined within the glass jar, either could have moved out of the tree and web. In the field later in the season I have often found an adult male in the web together with a female and her egg-cases. It is also not rare to find more than two spiders in a web. A casual search of perhaps 30-40 occupied webs during two days in April 1976 showed six of them each with three spiders: four with a female and two males, two with two females and a male. DISCUSSION Comparison of sexual behavior in Dictyna volucripes with what is known from other dictynids shows this species to be quite generalized for the family. Its pattern of courtship and mating is especially close to that described by Billaudelle (1957) from D. civica (H. Luc.). Each of the behaviors recorded from D. volucripes appears to occur in at least one other species. Among the generalized features of D. volucripes sexual behavior are: twitching of the male’s abdomen during spinning, face-to-face approach and stroking, mating position I, insertion of one palp at a time, alternation of palps in subsequent insertions, tendency to remain together for some days after mating, and the overall lack of aggression within the pair. I have assumed that abdominal twitching and face-to-face contact are each homologous in different species, though Jackson (1979) noted differences in form. STARR — DICTYNA SEXUAL BEHAVIOR 327 FEMALE MALE at rest, introduced at base usually in retreat climb onto tree Fig. 5. — Diagram of observed sexual behavioral sequences of D. volucripes. Dashed lines delimit sexual behavior. The sequences of female and male behaviors in D. volucripes are shown in Fig. 5. Jackson (1979) divided courtship in dictynids into a non-contact and a contact phase. This division is evident in D. volucripes , though I prefer to distinguish three phases. Ranging-spinning is purely a non-contact phase, local-spinning is a transition phase, and the stroking phase is purely a contact phase. Sperm induction evidently takes place before the start of courtship, as it was not observed in any trial. It appears usual for dictynids to re-induce sperm soon 328 THE JOURNAL OF ARACHNOLOGY after mating (Gerhardt and Kaestner 1937; Billaudelle 1957; Bristowe 1971), but we did not see this in D. volucripes. The present results are consistent with Jackson’s (1979) conclusion that vision has little or no role in dietynid courtship. The finding that males on recently vacated webs courted normally in the ranging-spinning and first part of the local- spinning phases likewise corroborates his conclusion that it is the female’s silk which releases and directs courtship in his “non-contact phase”. Any study of courtship in spiders suffers from the burden that its principal function is not yet established. Despite decades of controversy (for a summary review see Robinson and Robinson 1980), the two main contending hypotheses remain the same; Successful courtship (a) inhibits a very predatory animal (the female) from attacking a very edible one (the male), or (b) stimulates the female to accept the male as a mate. The two hypotheses need not be mutually exclusive, but the question remains of which is the limiting factor in courtship evolution. The predation-inhibition hypothesis is so attractive that it long had near hegemony among araneologists. For a recent explicit example of this view, see Gertsch (1979). T. H. Savory’s repeated protest (e.g., Savory 1928) that sexual approach is in fact rarely hazardous for male spiders seems to have had little impact, possibly because other aspects of his view of courtship are so hard to accept. Recent studies (e.g., Jackson 1979; Robinson and Robinson 1980), however, increasingly support the view that courtship is mainly a matter of female mate-choice. That is to say, it requires little effort to inhibit the female’s predatory drive, but much to gain acceptance as a mate. On a larger scale, this is in line with the view of animal courtship as shaped mainly by female choice and not by a need for species-recognition (Thornhill and Alcock 1983; West-Eberhard 1984; Eberhard 1985). The present study was not made with either hypothesis in mind, but I believe it contributes to this question. I interpret the results as much more consistent with female choice than with a need to inhibit predation. Let me mention in passing that I reach this conclusion reluctantly, as the predation-inhibition hypothesis has always for me invested spider and scorpion courtship with special fascination. In none of the 13 trials was there any indication that the male was in serious danger. Only in a minority of trials did the female rash at or otherwise vigorously approach him, and in each case he easily retreated out of reach. On the other hand, there were good indications of female choice. In three of 13 trials, normal courtship failed to lead even to a first insertion. In two of eight trials in which the male courted beyond the first insertion, he did not achieve a second insertion. There are further indications of mate-choice in the behavior of females. Females showed much less behavioral variety than males (Fig. 5) and after the initial reaction during ranging-spinning they mostly remained passive in the retreat. In a few cases the female retreated slightly from the male during local-spinning or stroking, apparently in resistance to courtship. Mate-choice is also implied in raising the female prior to insertion. In the Results and Fig. 5, I treat this as an active process on the male’s part and passive on the female’s part. Although I cannot be certain of this, it had that appearance, and the first two pairs of legs are surely strong enough to raise another spider’s body. At the same time, it seems dear that a female which grasps the substrate silk with flexed legs could not be lifted by force, and females sometimes appeared to resist in this way. In some trials the male was seen to repeatedly reach his legs STARR — DICTYNA SEXUAL BEHAVIOR 329 Table I. — Period of hematodocha pulsation in dictynid spiders. Explanation in text. Billaudelle (1957) in fact specified 2-3 pulsations/ s in D. civica , but I assume he meant one per 2-3 s. SPECIES PERIOD (seconds) REFERENCE Dictyna benigna about 4-30 (increasing during time of insertion) Karpinski 1882 Dictyna civica 2-3 Billaudelle 1957 Dictyna sublata about 6 Montgomery 1903 Dictyna volucripes 2-10 (mostly 7-9) this paper Heterodictyna viridissima about 10 Berland 1916 under the female’s carapace in apparent unsuccessful attempts to raise her. The best interpretation of stroking, then, is that it serves to overcome resistance to raising. The tendency to revert to courtship between insertions is part of the usual pattern in dictynids (Jackson 1979). The general lack of female aggression and her almost complete passivity at this time make it hard to reconcile such renewed courtship with any need to inhibit predation. Jackson (1979) has reviewed the durations of insertions reported from dictynids. These are almost all between 15 min and 2 h, much like those recorded from D. volucripes. Pulsations of the hematodocha during mating have previously been timed in four species (Table 1). These mostly have a period of 2- 10 sec, likewise in the range recorded in D. volucripes. Prolonged cohabitation has been reported from several families of spiders, but seems especially prevalent in the Dictynidae (Bristowe 1971; Gertsch 1979; Jackson 1979). Together with the generally nonaggressive nature of sexual activity, this led Bristowe (1971) to remark on “the unusual friendship which seems to exist between the males and females” in this family. The adaptive basis for such cohabitation is unknown. At present the best hypothesis seems to be that it functions primarily in mate-guarding by males Jackson (1977). If the function of cohabitation is obscure, its main social-evolutionary implication seems clear. Any mechanism which facilitates mutual tolerance among conspecifics removes a key obstacle to sociality. Cohabitation cannot explain why some dictynids are social, but it shows why they need not be solitary. ACKNOWLEDGMENTS In my first paper on arachnids I deem it suitable to acknowledge my teachers in the subject, C. D. Dondale and J. H. Redner of Canada Agriculture and R. E. Beer of the University of Kansas. This study began as a project in Dr. Beer’s Arachnology class and has subsequently benefitted from advice and criticism by C. D. Dondale, W. J. Gertsch, R. R. Jackson and M. H. Robinson. S. Pierce assisted in collecting data. I also thank L. Moortgat for statistical advice, M. M. Starr for volunteer typing and G. Venable for preparing Fig. 5. LITERATURE CITED Berland, J. 1916. Note preliminaire sur le cribellum et le calamistrum des Araignees cribellates et sur les moeurs de ces Araignees. Arch. Zool. Exp., 55:53-66. (Cited from Chamberlin and Gertsch 1958). 330 THE JOURNAL OF ARACHNOLOGY Billaudelle, H. 1957. Zur Biologie der Mauerspinne Dictyna civica (H. Luc.) (Dictynidae, Araneida). Z. angew. Entomol., 41:474-512. Bristowe, W. S. 1971. The World of Spiders. 2nd edition. Collins, London. 304 pp. Chamberlin, R. V. and W. J. Gertsch. 1958. The spider family Dictynidae in America north of Mexico. Bull. American Mus. Nat. Hist., 116:1-152. Eberhard, W. G. 1985. Sexual selection and animal genitalia. Harvard Univ. Press, Cambridge. 244 PP- Gerhardt, U. 1924. Weitere Studien fiber die Biologie der Spinnen. Arch. Naturges., 90:85-192. Gerhardt, U. and A. Kaestner. 1937. Araneae = Echte Spinnen = Webspinnen. Pp. 394-496, In Handbuch der Zoologie, (W. Ktikenthal and T. Krumbach, eds.), vol. 3. DeGruyter, Berlin. Gertsch, W. J. 1979. American Spiders. 2nd edition. Van Nostrand Reinhold, New York. 274 pp. Honjo, S. 1977. Social behavior of Dictyna foliicola Bos. et Str. (Araneae: Dictynidae). Acta Arachnol, 27:213-219. Jackson, R. R. 1977. Web sharing by males and females of dictynid spiders. Bull. British Arachnol. Soc., 4:109-112. Jackson, R. R. 1978a. Male mating strategies of dictynid spiders with differing types of social organization. Symp. Zool. Soc. London, 42:74-88. Jackson, R. R. 1978b. Comparative studies of Dictyna and Mallos (Araneae, Dictynidae). I. Social organization and web characteristics. Rev. Arachnol., 1:133-164. Jackson, R. R. 1979. Comparative studies of Dictyna and Mallos (Araneae, Dictynidae). II. The relationship between courtship, mating, aggression and cannibalism in species with differing types of social organization. Rev. Arachnol., 2:103-132. Karpinski, A. 1882. Ueber den Bau des mannlichen Tasters und den Mechanismus der Begattung bei Dictyna benigna Walck. Biol. Centralbl., 1:710-715. Leech, R. E. 1966. The spiders (Araneida) of Hazen Camp 81 49'N, 71 18'W. Quaest. Entomol., 2:153- 212. Locket, G. H. 1926. Mating habits of some web-spinning spiders, with some corroborative notes by W. S. Bristowe. Proc. Zool. Soc. London, 2:1125-1146. Montgomery, T. H. 1903. Studies of the habits of spiders, particularly those of the mating period. Proc. Acad. Nat. Sci. Philadelphia, 55:59-149. Robinson, M. H. and B. Robinson. 1980. Comparative studies of the courtship and mating behavior of tropical araneid spiders. Pacific Ins. Monogr. no. 36, 218 pp. Savory, T. H. 1928. The Biology of Spiders. Sidgwick & Jackson, London. 376 pp. Scheffer, T. H. 1905. The cocooning habits of spiders. Kansas Univ. Sci. Bull., 3:85-120. Thornhill, R. and J. Alcock. 1983. The evolution of insect mating systems. Harvard Univ., Cambridge. 547 pp. Uetz, G. W. 1983. Sociable spiders. Nat. Hist., 92:62-69. West-Eberhard, M. J. 1984. Sexual selection, competitive communication and species-specific signals in insects. Symp. R. Ent. Soc. London no. 12:283-324. Manuscript received February 1988, revised April 1988. Oraze, M. J., A. A. Grigarick, J. H. Lynch and K. A. Smith. 1988. Spider fauna of flooded rice fields in northern California. J. Arachnoh, 16:331-337. SPIDER FAUNA OF FLOODED RICE FIELDS IN NORTHERN CALIFORNIA Michael J. Oraze, Albert A. Grigarick, Joseph H. Lynch and Kirk A. Smith Department of Entomology University of California Davis, California 95616 USA ABSTRACT A survey of the spiders associated with northern California rice fields was conducted to identify potential biological-control agents of rice feeding insects and mosquitoes. All of the 28 species were collected on the levees; however, only 10 of these were taken in the paddies. Pardosa ramulosa (McCook), Pirata piraticus (Clerck) and two linyphiid spp. were common throughout the agroecosystem. These spiders exhibited a seasonal succession in relative abundance within the paddy during the growing season. Pardosa ramulosa was dominant both on the levees and in the paddies. It comprised ca. 58 and 68% of the fauna in these respective areas. We suggest that the flooded paddies may serve as a refuge for the semiaquatic P. ramulosa during the dry summer months and that its abundance in California rice fields is due in part to the similarity of this agroecosystem to the native, pre-agricultural habitat. INTRODUCTION Rice, Oryza saliva L., was introduced into California in 1912 and is now grown annually on about 161,880 ha (400,000 acres). Rice production is a major industry in the Sacramento Valley where more than 90% of the rice acreage in the state is located. A rice field is a complex agroecosystem, containing many aquatic, semiaquatic, and terrestrial species. Spiders are well represented among the many predators found in this habitat. They feed mostly on insects and may contribute in reducing pest levels. Lower pest densities have been attributed to spider activity in rice fields of Asia (Kiritani 1979) and other agroecosystems worldwide (Rkchert and Lockley 1984). Numerous surveys of spiders have been conducted in the rice growing regions of Asia (Barrion and Litsinger 1984). However, little is known about spiders associated with rice in the United States. Preliminary surveys have been conducted in Texas (Woods and Harrell 1976) and Arkansas (Heiss and Meisch 1985), but no attempt has yet been made to formally describe the California rice field fauna. This paper identifies the spiders collected from the levees and flooded paddies of several California rice fields over a three year period. MATERIALS AND METHODS Sampling began in 1983 at the following locations: the Beck ranch near Modesto (Stanislaus County), Van Dyke ranch near Natomas (Sutter County), 332 THE JOURNAL OF ARACHNOLOGY and in the Lattemore seed-field section of the Rice Expriment Station near Biggs (Butte County). Sampling efforts in 1984-1985 were limited to the Biggs site after it was determined the three areas yielded nearly identical results with respect to common species. The California rice field habitat and associated common vegetation were described by Barrett and Seaman (1980). Notable differences in vegetation among sampling sites used in this study included dense populations of common cattail ( Typha latifolia L.) and bermuda grass ( Cynodon dactylon (L.) Pers.) on the levees at the Modesto and Natomas sites, respectively, but not at Biggs. Monochoria ( Monochoria vaginalis (Burm. f.) Presl.) and toothcup ( Rotala indica (Willd.) Koehne) were restricted to and abundant in the paddies at the Rice Experiment Station. Wide mouth Mason jars (11 cm deep and 7.5 cm in diameter) served as pitfall traps. They were inserted into plastic sleeves that were permanently buried in the levees flush with the soil surface. Ten traps were installed at each site at ca. 8 m intervals in an alternating pattern (north side, center, south side, etc.) along the length of a levee in selected fields. One hundred and fifty ml of 95% ethylene glycol plus 5% liquid detergent was added to each trap. After seven days, the traps were collected. The contents were filtered through a USA Standard Testing Sieve No. 40 and stored in 70% EtOH. This procedure was repeated monthly throughout the growing season (May-September). Floating sticky traps were made by cutting white styrofoam into triangular wedges 61 cm long, 4.5 cm high, with bases of 9 cm. A thin coat of Stickem Special™ (Seabright Enterprises Ltd.; Emeryville, CA) was brushed on the upper surfaces. Five traps were placed in each field and positioned equidistant from one another (ca. 34 to 92 m apart depending on field size) along a transect connecting the NE and SW corners of the field with the end traps being placed 2 m from the margins. They were held in place with green bamboo stakes in a manner that allowed the traps to move vertically so contact with the fluctuating water surface could be maintained. The stakes also served to mark the position of the traps. After seven days, the traps were collected and the spiders identified (to species when possible) with the aid of a 10X hand lens. The sampling schedule was the same as that for the pitfall traps. Companion samples were taken in 1983 with a UC-VAC® suction device (Summers et al. 1984) to estimate absolute densities and determine the nature and extent of any bias associated with pitfall and sticky-trap sampling. Ten samples ca. 10 m apart were collected both on the levees and in the paddies. A circular unit-area-sampler, enclosing 0.093 m2 (1 ft.2) and standing 38 cm (15 in.) high, was placed in the general vicinity of the pitfall and floating sticky traps. The enclosed substrate and vegetation were vacuumed for ca. 90 s. Sampling was conducted between 1200 and 1400 hours. Samples were immediately placed in a cooler with ice for transport, and later processed in Berlese funnels for 48 h. RESULTS AND DISCUSSION More than 30,000 specimens were collected in the survey. Species that were taken at all sampling sites in every year — representing 11 families, 22 genera and 28 species — are listed in Table 1. They have been ranked as 4 common, 2 ORAZE, ET AL.— SPIDERS IN CALIFORNIA RICE FIELDS 333 Table 1. — Spiders collected in northern California rice fields (1983-85). a = L, levee; P, paddy, b R = rare (< 1%); O = occasional (1-5%); C = common (> 5%). Species frequencies determined by averaging counts from UC-VAC (levee) and pitfall-trap samples. Taxa Location3 Frequency1* Dysderidae Dysdera crocata C. L. Koch L R Linyphiidae Erigoninae Species A L,P C Species B L,P C Araneidae Araneus trifolium (Hentz) L,P R Argiope aurantia Lucas L,P R Argiope trifasciata (Forskal) L,P R Tetragnathidae Tetragnatha elongata Walckenaer L,P R Tetragnatha laboriosa Hentz L,P R Lycosidae Alopecosa kochi (Keyserling) L R Pardosa ramulosa (McCook) L,P C Pirata piraticus (Clerck) L,P C Oxyopidae Oxyopes salticus Hentz L R Gnaphosidae Drassyllus insularis (Banks) L R Drassyllus saphes Chamberlin L R Micaria sp. L R Trachyzelotes lyonneti (Audouin) L R JJrozelotes rusticus (L. Koch) L R Zelotes puritanus Chamberlin L R Thomisidae Xysticus californicus Keyserling L R Philodromidae Tibellus oblongus (Walckenaer) L,P O Salticidae Habronattus klauserii (Peckham & Peckham) L R Metaphidippus vitis (Cockerell) L R Neon ellamae Gertsch & Ivie L R Phidippus californicus Peckham & Peckham L R Phidippus clarus Keyserling L R Phidippus johnsoni (Peckham & Peckham) L R Sitticus dorsatus (Banks) L R Dictynidae Tricholathys saltona Chamberlin L O occasional and 22 rare species. All 28 species were collected on the levees (including vegetation) but, only 10 of the species were taken in the paddy. Apparently many of the levee species were incapable of inhabiting an aquatic microhabitat. Only four species — Pardosa ramulosa (McCook), Pirata piraticus (Clerck) and two linyphiid spp. — were common in the paddy. Other spider species occasionally found in the paddy were generally limited to the paddy margins late in the growing season after the crop canopy had filled in enough to allow plant- to-plant movement and provide adequate sites for web attachment. The four common paddy spiders also dominated the levee fauna. The number of taxa recorded are generally lower than those reported for other surveys (Paik and Kim 1973). This can be attributed in part to our exclusion of 334 THE JOURNAL OF ARACHNOLOGY Table 2. — Relative abundance in percent composition of the major spiders in northern California rice fields, a = 1983; b = Pitfall or sticky-trap catches from 1983-1985. Taxa Levee Paddy UC-VACa (n = 1,099) Pitfall15 (n = 16,311) UC-VAC (n = 614) Sticky*5 (n = 12,124) Linyphiid spp. 12.9 3.4 19.3 5.1 Pirata piraticus 8.4 4.0 11.4 6.4 Pardosa ramulosa 57.7 75.2 67.5 87.4 Others 21.0 17.4 1.8 1.1 transient species (species not collected at every sampling site in every year, usually represented by a single specimen). Even so, comparisons of this type may be misleading, as great differences exist among the surveys in terms of the extent and methods of sampling. For example, Woods and Harrell (1976) collected 752 specimens from a single 14.8 ha (37 acre) field during one growing season. In contrast, Barrion and Litsinger (1984) collected 13,270 specimens from 17 localities over three years, and Okuma (1968) collected 1,487 spiders from 22 localities during a 10 day period. Furthermore, Heiss and Meisch (1985) sampled with an aquatic net and metal dipper but Okuma and Wongsiri (1973) utilized a sweep net and observations. In spite of these differences, three families: Araneidae, including Tetragnathi- dae; Linyphiidae, including the Erigoninae (Micryphantidae) and Lycosidae dominated the spider fauna in all but one of the surveys of rice fields cited in this paper. In addition, the relative abundances of these families changed with latitude. In semitropical rice-growing areas, such as Taiwan, Thailand and the Philippines, araneids dominated (Okuma 1968; Chu and Okuma 1970; Okuma and Wongsiri 1973; Barrion and Litsinger 1984) while lycosids were more abundant in temperate regions such as Korea and the United States (Paik and Kim 1973; Woods and Harrell 1976; Heiss and Meisch 1985; present study). Lycosids and araneids were also abundant in the rice fields surveyed in Japan, although the fauna was dominated by two theridiid spp. (Paik and Kim 1973). Pardosa ramulosa was dominant in numbers on the levees and in the paddies (Table 2). It comprised ca. 58 and 68% (UC-VAC samples) of the fauna in these respective areas. The two lycosids, Pardosa ramulosa and Pirata piraticus , together constituted ca. 80% (UC-VAC samples) of the paddy spiders. They appeared to be well adapted to the water surface where they quickly ran about or remained motionless for long periods. They occasionally went underwater by crawling down emergent vegetation or debris. Other paddy spiders, although capable of limited locomotion on the water surface, spent most of their time on vegetation or in webs constructed among the paddy plants. Linyphiids were also seasonally abundant. However, their contribution to total spider biomass over the growing season was relatively small, compared to that of the common lycosids, because of their small size and ephemeral occurrence. Sticky and pitfall-trap samples probably overestimated Pardosa ramulosa while underestimated linyphiids and Pirata piraticus abundances compared to UC-VAC samples (Table 2). Although fewer spiders were collected with the UC-VAC, these data were probably more accurate in estimating relative abundances for the major species. The UC-VAC was not used more extensively because of the much greater GRAZE, ET AL.— SPIDERS IN CALIFORNIA RICE FIELDS 335 I I linyphiids | Pardosa ramulosa Pirata piraticus May June July August Sept. Fig. 1. Relative seasonal abundance of the major paddy spiders in northern California rice fields for 1984 (sticky-trap catches). relative time and effort it required. Foliage dwelling species such as Tibellus oblongus (Walckenaer) were only collected with the UC-VAC whereas nocturnal ground dwellers such as the dictynids and gnaphosids were limited to pitfall traps. This illustrates the importance of utilizing multiple collecting techniques in faunal surveys of spiders. The major paddy spider species exhibited a seasonal succession in relative abundance during the growing season (Fig. 1). The linyphiids dominated the spider fauna in the paddies shortly after flooding. Their abundance was associated with the spring ballooning period when they arrived in massive numbers. Unlike the other two major paddy species, the linyphiids do not appear to be specifically adapted for, or restricted to, aquatic environments. Pirata piraticus is distributed throughout Europe and north of the 35th parallel in North America. It is associated with swamps, marshes and the shores of lakes, ponds and streams (Wallace and Exline 1977). It became a major component of the paddy fauna late in the growing season. Pardosa ramulosa is found throughout California. Its range extends E through southern Nevada into the SW corner of Utah and S into northern Mexico. In California it is one of the dominant lycosids at elevations below 300 m (Hydorn 1977). It is associated with mesic habitats such as salt marshes (Garcia and Schlinger 1972; Greenstone 1980), sewage oxidation ponds (Hydorn 1977), irrigated lawns (Van Dyke and Lowrie 1975) and irrigated crops (Leigh and Hunter 1969; Yeargan and Dondale 1974; Hickle 1981). The prevalence of P. ramulosa in rice and other irrigated crops in California is probably related to the seasonal compression of suitable habitat. As drying begins in the spring and continues through the summer these spiders are probably forced to aggregate where moist conditions persist. Irrigated cropland, particularly rice, which is typically continuously flooded from May through September, offers such a refuge. Rice culture in California resembles the native habitat of some areas that existed before the advent of flood control and irrigation projects when many parts of the Sacramento and San Joaquin Valleys were annually flooded from snowmelt. Rice fields probably represent the functional equivalent of the numerous vernal ponds and marshes that were presumably utilized by P. ramulosa in its pristine environment, but differ by extending moisture availability, which is essential for this species (Hydorn 1977), throughout the summer. For a large part of the growing season, this native natural enemy is actually favored by, and more abundant in, rice (an introduced annual crop) than in adjacent untilled 336 THE JOURNAL OF ARACHNOLOGY border areas (Graze et al. unpublished data). Because of its abundance in, and preadaptation to, the rice field environment, Pardosa ramulosa appears to be the spider most likely to contribute to a level of biological control of one or more insect pests in this agroecosystem. The impact of this spider on selected prey species in rice will be presented in a subsequent paper (Oraze and Grigarick 1988). Our sampling did not include any wild rice ( Zizania aquatica L.) fields. This crop supports more vegetative growth and is usually produced earlier in the year (February-July) than conventional rice. We suspect that these cultural differences may cause minor differences in the respective spider faunas, and a comparative study would be of value. ACKNOWLEDGMENTS We thank M. O. Way for suggesting the project; O. G. Bacon, W. Johnson, and B. Brandon for technical assistance; R. Stein for collecting and counting samples; B. Beck, R. Van Dyke, and the staff at the Rice Experiment Station for the use of their fields; E. I. Schlinger, J. A. Coddington, N. I. Platnick and G. B. Edwards for identifying spiders; R. Karban, D. Spiller, R. F. Chapman and L. E. Ehler for reviewing the manuscript; and G. B. Edwards for his helpful comments. This work was supported by grants from the Entomology Department Univ. of Calif., Davis and the California Rice Research Board. This paper is part of a dissertation submitted by M. J. O. in partial fulfillment of a Ph.D. requirement. LITERATURE CITED Barrett, S. C. H. and D. E. Seaman. 1980. The weed flora of Californian rice fields. Aquat. Bot., 9:351-376. Barrion, A. T. and J. A. Litsinger. 1984. The spider fauna of Philippine rice agroecosystems. II Wetland. Philippine Entomol., 6:11-37. Chu, Y. and C. Okuma. 1970. Preliminary survey on the spider-fauna of the paddy fields in Taiwan. Mushi, 44:65-88. Garcia, R. and E. I. Schlinger. 1972. Studies of spider predation on Aedes dorsalis (Meigen) in a salt marsh. Pp. 117-118, In Proceedings and Papers 40th Annual Conference California Mosquito Control Association. Greenstone, M. H. 1980. Contiguous allotopy of Pardosa ramulosa and Pardosa tuoba (Araneae: Lycosidae) in the San Francisco Bay Region, and its implications for patterns of resource partitioning in the genus. American Mid. Nat., 104:305-311. Heiss, J. S. and M. V. Meisch. 1985. Spiders (Araneae) associated with rice in Arkansas with notes on species composition of populations. Southw. Nat., 30:119-127. Hickle, L. A. 1981. The biology and ecology of Pardosa ramulosa (McCook), (Araneae, Lycosidae), in Imperial Valley, California. Ph.D. dissertation, Univ. of California, Riverside. Hydorn, S. E. 1977. The biology of Pardosa ramulosa. Ph.D. dissertation, Univ. of California, Berkeley. Kiritani, K. 1979. Pest management in rice. Annu. Rev. Entomol., 24:279-312. Leigh, T. F. and R. E. Hunter. 1969. Predaceous spiders in California cotton. California Agric., 23:4- 5. Okuma, C. 1968. Preliminary survey on the spider-fauna of the paddy fields in Thailand. Mushi, 42:89-117. Okuma, C. and T. Wongsiri. 1973. Second report on the spider-fauna of the paddy fields in Thailand. Mushi, 47:1-17. GRAZE, ET AL.— SPIDERS IN CALIFORNIA RICE FIELDS 337 Oraze, M. J. and A. A. Grigarick. 1988. Biological control of aster leafhopper (Homoptera: Cicadellidae) and midges (Diptera: Chironomidae) by Pardosa ramulosa (Araneae: Lycosidae) in California rice fields. J. Econ. EntomoL, in press. Paik, K. and J. Kim. 1973. Survey on the spider-fauna and their seasonal fluctuation in paddy fields of Taegu, Korea. Korean J. Plant Prot., 12:125-130. Riechert, S. E. and T. Lockley. 1984. Spiders as biological control agents. Annu. Rev. Entomol., 29:299-320. Summers, C. G., R. E. Garrett and F. G. Zalom. 1984. New suction device for sampling arthropod populations. J. Econ. Entomol. 77:817-823. Van Dyke, D. and D. C. Lowrie. 1975. Comparative life histories of the wolf spiders Pardosa ramulosa and P. sierra (Araneae: Lycosidae). Southw. Nat., 20:29-44. Wallace, H. K. and H. Exline. 1977. Spiders of the genus Pirata in North America, Central America and the West Indies (Araneae: Lycosidae). J. Arachnol., 5:1-112. Woods, M. W. and R. C. Harrell. 1976. Spider populations of a southeast Texas rice field. Southw. Nat., 21:37-48. Yeargan, K. V. and C. D. Dondale. 1974. The spider fauna of alfalfa fields in northern California. Ann. Entomol. Soc. America, 67:681-682. Manuscript received March 1988, revised April 1988. Beatty, J. A. and J. W. Berry. 1988. Four new species of Parathewna (Araneae, Desidae) from the Pacific. J. Arachnol., 16:339-347. FOUR NEW SPECIES OF PARATHEUMA (ARANEAE, DESIDAE) FROM THE PACIFIC Joseph A. Beatty Department of Zoology Southern Illinois University Carbondale, Illinois 62901 USA and James W. Berry Department of Biological Sciences Butler University Indianapolis, Indiana 46208 USA ABSTRACT Four new species of Parathewna are described from the Cook, Fiji and Tuamotu Islands and Australia. New records of P. armata are presented. Possible relationships among species of the genus are suggested. INTRODUCTION In a recent paper (Beatty and Berry 1988) we illustrated and discussed the three known species of the genus Paratheuma Bryant. Here we report the rather surprising subsequent discovery of four new species of the genus from Australia and the Cook, Fiji and Tuamotu Islands of the South Pacific. Like the other species, these were taken near the high tide level on seashores, often among loose broken coral thrown up on the beach. However, we found some individuals under large non-coralline rocks, and others in crevices or holes in outcrops of volcanic or conglomerate rocks a few cm to about one meter above normal high tide level. The previously known species of Paratheuma are P insulana (Banks) from Bermuda, Florida, Cuba and Haiti (Banks 1902, 1903; Beatty and Berry 1988; Bryant 1940; Platnick 1977), P. inter aesta (Roth and Brown) from the northern part of the Gulf of California (Beatty and Berry 1988; Platnick 1977; Roth and Brown 1975), and P. armata (Marples) from Swains Island, Marshall Islands and Caroline Islands in the Pacific (Beatty and Berry 1988; Marples 1964). The four new species are quite similar to these in size, shape, coloration and setation, as well as habitat. We have already described (Beatty and Berry 1988) the range of coloration in the genus, and the additional species add little to this range. A few of the recently collected specimens, almost black on the abdomen, are darker than any we had 340 THE JOURNAL OF ARACHNOLOGY seen earlier. The light abdominal chevrons occasionally present are more distinct in some specimens than our previous description suggests. Size and proportions of all species show so little variation that we have presented only a few measurements in the descriptions below. Three adult males and three adult females of each species were measured, except for P. andromeda , of which we had only one male. We have not described the bristle pattern for each species individually, largely because the bristles are weak, not very abundant, and vary little among species. Instead, a separate description is presented, which applies equally well to all the Pacific species. These “bristles” are, of course, setae, but the presence of three main size classes of setae in spiders makes retention of the commonly used terms, “hairs, bristles and spines” useful for distinguishing among them. DESCRIPTIONS Setation. — All five of the Pacific Paratheuma have the same arrangement of bristles, with no more variation among the species than within a single population of one of them. There is almost no difference in the pattern between males and females. In the following description a bristle number indicated as 1-3 means one to three bristles; 1-2-3 means one proximal bristle, two near mid-length, and three distal, on a particular appendage surface. Palp. — Two dorsal bristles on femur, in distal half; two dorsal on patella, one proximal, one distal; two prolateral and one dorsal on tibia; on tarsus, two prolateral and two retrolateral near base (the retrolateral pair absent in adult males), a pair just distal to mid-length of tarsus, one on each side, another such pair at distal end, and two distal ventral bristles in the mid-line. Legs. — Femora'. In both sexes on all legs, 1-3 dorsal, 1 dorsolateral. Dorsolateral bristle dorsoprolateral on legs I-II, dorsoretrolateral on 1 1 I I V. Patellae: Two dorsal (one proximal, one distal) on all legs in both sexes. Tibiae: Leg I, 1-2 dorsal, 2-3 ventral. Leg II, 2 dorsal, 1 prolateral, 2 ventral. Leg III, 1- (0-2)-3 dorsal, 1-1-1 ventral. Leg IV, 1-2-3 dorsal, (l-2)-(l-2)-(l-3) ventral. Metatarsi: Legs I and II, 5-6 ventral arranged (l-2)-2-2. Leg III, (0-l)-2-2 dorsal, 2-2-3 ventral. Leg IV, 2-2-2 dorsal, 2-2-3 ventral. Juvenile specimens listed below are not to be regarded as paratypes. Holotype males and one paratype female of each species are deposited in the Bishop Museum, Honolulu. All other material examined remains in the possession of the authors. Paratheuma andromeda , new species Figs. 1-4 Holotype. — Male from Cook Islands, Aitutaki, Rapota Motu, in crevice in volcanic rock outcrop on shore, 5 June 1987 (J. W. Berry), in Bishop Museum, Honolulu, Hawaii, USA. The name andromeda is a noun in apposition after Andromeda of classical mythology. Diagnosis. — Male: The broad tibial apophysis of the male palp, curving dorsally and toward the cymbium (Fig. 4) clearly distinguishes andromeda from BEATTY & BERRY— PACIFIC PARATHEUMA 341 Figs. 1-8. — Left male pedipalp and epigynum of Paratheuma species: 1-4, P. andromeda from Aitutaki, Cook Islands; 1, pedipalp, ventral; 2, epigynum, ventral; 3, epigynum, cleared, dorsal; 4, tibia of pedipalp, lateral; 5-8, P. ramseyae from Rarotonga, Cook Islands; 5, tibia of pedipalp, lateral; 6, pedipalp, ventral; 7, epigynum, ventral; 8, epigynum, cleared, dorsal. all other species of the genus. The length and orientation of the distal processes of the conductor (Fig. 1) are also distinctive. Female: The large ovate seminal receptacles distinguish this species from all others except P. ramseyae. From the latter andromeda differs by having the entire 342 THE JOURNAL OF ARACHNOLOGY anterior margin of the epigynal depressions sclerotized, and by lacking the heavily sclerotized pouch around the epigynal openings (Figs. 2-3). Additional descriptive notes. — Male: Total length 3.8 mm, carapace length 1.7 mm, maximum carapace width 1.1 mm. Embolus originating at mid-length of bulb, curving around anteromedial margin of bulb, turning back to end in hairfine filament on conductor. Two rather long narrow distal processes of conductor almost parallel with long axis of tibia, curving slightly laterally (Fig. 1). Tibial apophysis of palp broad, curving strongly dorsally and toward cymbial base (Fig. 4). Female: Total length 4. 1-4.4 mm, carapace length 1.7-1. 8 mm, maximum carapace width 1.2-1. 3 mm. Epigynum with narrow sclerotized rim along entire anterior length of depressions, curving short distance along lateral margin. Broad heavily sclerotized ducts leading from openings to large ovate seminal receptacles. Distribution. — Known only from a small area of volcanic shoreline on one islet of Aitutaki, Cook Islands, and from the shore of a nearby islet. The apparently very restricted distribution of this species is unusual for the genus, though it is possible that it occurs on other main islands of the Cook group. A large percentage of the shoreline of the Aitutaki Islands was searched, without success, for additional specimens. Specimens examined. — COOK ISLANDS: Aitutaki ; Moturakau, in coral rubble on beach, 28 March 1987 (J. W. Berry), 1 female, 1 immature; Rapota Motu, in crevices and holes in volcanic and conglomerate rock at shoreline, 5 June 1987 (J. W. Berry), 1 male, 2 females. Paratheuma ramseyae, new species Figs. 5-8 Holotype. — Male from Cook Islands, Rarotonga, Koromiri Islet, in broken coral rubble on beach, 3 April 1987 (J. W. and E. R. Berry), in Bishop Museum, Honolulu, Hawaii, USA. The species is named after Elizabeth Ramsey Berry, who discovered it. Diagnosis. — Male: Distal end of conductor broad, with two short, bluntly pointed processes that curve slightly toward tibia (Fig. 6). Tibial apophysis slender, curving dorsally and toward cymbium (Fig. 5). Female: Distinguished from all other species by the short anterolateral sclerotizations of the epigynal depressions, and the large hoodlike sclerotized pouches around the epigynal openings (Figs. 7-8). Additional descriptive notes. — Male: Total length 3. 3-3. 9 mm, carapace length 1.6- 1.8 mm, maximum carapace width 1.2-1. 3 mm. Embolus of palp as in other species described here. Other palpal characters as in diagnosis. Female: Total length 3. 6-4. 5 mm, carapace length 1 .7-1.8 mm, maximum carapace width 1.3 mm. Sclerotizations of rim of epigynal openings short and sigmoid, located in anterolateral portions of depressions only. Openings leading into large sclerotized pouches. Short ducts from pouches to large ovate seminal receptacles. Distribution. — Cook Islands, Rarotonga. Known from the main island of Rarotonga itself, and from four small islets inside the fringing reef. Specimens examined. — COOK ISLANDS: Rarotonga ; Ngatangiia Harbor beach, in rock outcrops, 31 March 1987 (J. W. Berry and J. A. Beatty), 1 female, 1 immature; Koromiri Islet, in broken coral rubble on sand beach, 3 April 1987 (J. W. Berry and E. R. Berry), 2 males, 1 female, 8 immature; BEATTY & BERRY— PACIFIC PARATHEUMA 343 Koromiri Islet, in coral rubble, 4 April 1987 (J. W. and E. R. Berry), 1 male, 1 female, 5 immature; Koromiri Islet, in coral rabble, 6 April 1987 (3. W. and E. R. Berry ), 3 females, 28 immature; Oneroa Islet, in beach litter, 21 March 1987 (J. W. and E. R. Berry), 1 female, 2 immature; Taakoka Islet, in crevices on volcanic rock outcrop at shore, 19 March 1987 (J. W. Berry and J. A. Beatty), 1 male, 2 females; on offshore islets at Muri Beach, 27 March-1 April 1987 (J. W. and E. R. Berry), 3 males, 1 female, 3 immature. Paratheuma australis , new species Figs. 9-12 Holotypej — Male from Fiji, Viti Leva, Korotongo village, shoreline at Reef Resort, in coral rubble, 21 May 1987 (J. W. Berry and E. R. Berry), in Bishop Museum, Honolulu, Hawaii USA. The name australis is an adjective based upon the more southern range of this species, compared with most others in the genus, and its occurrence in Australia. Diagnosis. — Male: Palp with slender erect labial apophysis of medium length, as in P insulana and P rangiroa. Distinguished from insulana by its broader cymbium and palpal bulb, from rangiroa by the more anteriorly directed tibial apophysis and the shape of the conductor tip. Female: With narrow oblique epigynal depressions, as in P insulana and P. rangiroa . Distinguished from insulana by reduction of the anterolateral hoods of the epigynal depressions. See discussion under P. rangiroa for differences from that species. Additional descriptive notes. — Male: Total length 3. 2-3. 5 mm, carapace length 1.5- 1.7 mm., maximum carapace width 1.1 mm. Embolus originating on medial side of bulb near mid-point of bulb’s length, slender, tapering, curving in broad parabola to end in filament lying under edge of conductor, Tibial apophysis about 1 /5th length of cymbium, directed forward at slight angle to axis of tibia, slender and curving slightly dorsaiiy (Fig. 12). End of conductor broad, extended into two angular processes, one directed backward and slightly medially, one longer, extending laterally (Fig. 9). Female: Total length 3.4-4. 2 mm, carapace length 1.5-1. 7 mm, maximum carapace width 1.0-1. 1 mm. Epigynal depressions pale, extending obliquely forward and laterally, with only short narrow sclerotized rim anterolaterally. Openings at anterior medial edge of depressions, leading to short looped ducts ending posteriorly in pair of small receptacles (Figs. 10-11). Distribution. — Fiji Islands and Australia. Specimens examined. — AUSTRALIA: QUEENSLAND: Great Keppel Island, Leek's Beach, in coral rubble, 24 April 1987 (J. W. and E. R. Berry), 2 males; Monkey Beach, in coral rubble, 24 April 1987 (J. W. Berry), 4 immature; Yeppoon, Wave Point, among granitic rocks on beach, 23 April 1987 (J. W. and E. R. Berry), 1 male, 2 females, 7 immature. FIJI: Viti Levu; Nadi, Nadi Bay, in beach rabble, 29 April 1987 (J. W. and E. R. Berry), 1 male, 1 female; Nadi Bay, in gravel on beach (J. W. and E. R. Berry), 1 male, 4 immature; 2 km W of Vatukarasa, on beach among small coral rocks, 12 May 1987 (J. W. Berry), 1 female; Korotongo village, shore at Reef Resort, in coral rabble, 21 May 1987 (J. W. and E. R. Berry), 2 males, 2 females, 6 immature; 0.5 km E of Komave village, in coral rubble on beach, 24 May 1987 (J. W, and E. R. Berry), 1 male, 2 females, 21 immature. 344 THE JOURNAL OF ARACHNOLOGY Figs. 9-16. — Left male pedipalp and epigynum of Paratheuma species: 9-12, P. australis from Viti Levu, Fiji Islands; 9 pedipalp, ventral; 10, epigynum, ventral; 11, epigynum, dorsal, cleared; 12, tibia of pedipalp, lateral; 13-16, P. rangiroa from Rangiroa, Tuamotu Islands; 13, tibia of pedipalp, lateral; 14, pedipalp, ventral; 15, epigynum, ventral; 16, epigynum, cleared, dorsal. Paratheuma rangiroa , new species Figs. 13-16 Holotype. — Male from Tuamotu Islands, Manihi, Topihairi Islet, in beach rubble, 3 June 1987 (E. R. Berry), in Bishop Museum, Honolulu. The name is a noun in apposition after the atoll where the species was first found. Diagnosis. — Male: Palp with slender erect tibial apophysis of medium length, as in P. insulana and P. australis. Distinguished from insulana by its broader BEATTY & BERRY— PACIFIC PARATHEUMA 345 cymbium and palpal bulb, from australis by the more laterally directed tibial apophysis and the shape of the conductor tip, the medial projection of which is smaller and the lateral projection longer than in australis. Female: With narrow oblique epigynal depressions as in P. insulana and P australis. Distinguished from insulana by reduction of the anterolateral hoods of the epigynal depressions, from australis by the longer sclerotized rim of the depressions and the somewhat larger and more oblong seminal receptacles. Discussion: This species is genitalically quite similar to P australis. It is smaller than australis , and the genitalic differences, though slight, are constant in the available material. The presence of three more quite distinct species of the genus on islands between the ranges of rangiroa and australis argues for their distinctness. Comparison of these two species by electrophoresis of proteins from whole body extracts (Laemmli 1970, for method) showed clear differences. Additional descriptive notes. — Male: Total length 2. 7-2. 8 mm, carapace length 1.3- 1.4 mm, maximum carapace width 0.9- 1.0 mm. Palp as in P. australis except for differences noted in diagnosis (Figs. 13-14). Female: Total length 3. 1-3.7 mm, carapace length 1.4- 1.6 mm, maximum carapace width 1.0-1. 1 mm. Epigynum as in P. australis except for differences noted in diagnosis (Figs. 15-16). Distribution. — Known only from Rangiroa and Manihi Atolls in the Tuamotu Islands of French Polynesia. Specimens examined. — FRENCH POLYNESIA: Tuamotu Islands; Rangiroa Atoll, on lagoon beach near airport, 17 January 1987 (J. W. and E. R. Berry), 1 female, 1 immature; Avitorua Islet, in beach rubble’ 5 June 1987 (E. R. Berry), 1 male, 10 females, 18 immature; Manihi, Topihairi Islet, 3-4 June 1987 (E. R. Berry), 6 males, 3 females, 23 immature. Paratheuma armata (Marples) New records.— FRENCH POLYNESIA: SOCIETY ISLANDS; Moorea, Paopao village, in intertidal coral rubble, 10 January 1987 (J. W. Berry), 1 male, 1 female, 5 immature; Paopao village, in coral rubble on beach (E. R. Berry), 2 males, 6 females; Paopao village, beach rubble, 19 February 1987 (J. W. and E. R. Berry), 2 males, 4 immature; Paopao village, beach rubble, 22 February 1987 (J. W. Berry), 1 male, 2 females; west side of Moorea, in pile of supratidal rocks along shore, 13 January 1987 (J. W. and E. R. Berry), 2 males, 7 immature; Tahiti , Faaa, in coral rubble on beach, 20 February 1987 (J. W. and E. R. Berry), 7 immature; Faaa, in coral rubble on beach, 9 June 1987 (J. W. and E. R. Berry), 1 male. SPECIES RELATIONSHIPS We are not familiar enough with other members of the family Desidae to select a genus as the probable nearest relative of Paratheuma. (Also, the contents of the family seem not clearly determined at present). Consequently we do not feel justified in identifying specific character states as primitive or derived. Within the genus, australis and rangiroa appear to be the most closely related pair of species, and both are rather similar to insulana from the Caribbean. This suggests a vicariant relationship between insulana and the pair of Pacific species. As judged by female genitalia, andromeda and ramseyae seem close to each other, as would be expected by their geographic proximity. The other two species, armata and interaesta, are more distinctive and do not appear to have any close relatives among the known species of the genus. 346 THE JOURNAL OF ARACHNOLOGY Fig. 17. — Distribution of Paratheuma species in the Pacific. 17 The distributions of all known Pacific species of Paratheuma are entirely on or extend only slightly beyond the borders of the Pacific crustal plate (Fig. 17). Given the usually accepted view that all the islands on this plate have always been highly isolated, dispersal must have been of primary importance in the evolutionary history of the Pacific Paratheuma. Judging from our recent experience, however, there is a strong possibility that several, perhaps many, more species of the genus await discovery, and their distribution and characters may change our interpretation of their relationships and history considerably. ACKNOWLEDGMENTS Preparation of this paper was supported by a grant from Butler University, Indianapolis, Indiana. We are indebted to Elizabeth Ramsey Berry for her contributions to many aspects of the research. The staff of the Bishop Museum in Honoluly gave us access to their collection and a work space, as well as assisting in other ways, for which we are grateful. LITERATURE CITED Banks, N. 1902. Some spiders and mites from the Bermuda Islands. Trans. Connecticut Acad. Arts Sci., 11:267-275. BEATTY & BERRY— PACIFIC PARATHEUMA 347 Banks, N. 1903. A list of Arachnida from Hayti, with descriptions of new species. Proc. Acad. Nat. Sci. Philadelphia, 55:340=345. Beatty, J. A., and J. W. Berry. 1988. The spider genus Paratheuma (Araneae, Desidae). J. Arachnol., 16:47-54. Bryant, E. B. 1940. Cuban spiders in the Museum of Comparative Zoology. Bull. Mus. Comp. Zool., 86:249-532. Laemmli, U. K. 1970. Cleavage and structural proteins during the assembly of the head of bacteriophage T4. Nature, 227:680. Marples, B. J. 1964. Spiders from some Pacific Islands, Part V. Pacific Sci., 18:399-410. Platnick, N. I. 1977. Notes on the spider genus Paratheuma Bryant (Arachnida, Araneae). J. Arachnol., 3:199-201. Roth, V. D., and W. L. Brown. 1975. A new genus of Mexican intertidal zone spider (Desidae) with biological and behavioral notes. Amer. Mus. Novitates, 2568:1-7. Manuscript received October 1987, revised May 1988. Whitehouse, M. E. A. 1988. Factors influencing specificity and choice of host in Argyrodes antipodiana (Theridiidae, Araneae). J. Arachnol, 16:349-355. FACTORS INFLUENCING SPECIFICITY AND CHOICE OF HOST IN ARGYRODES ANTIPODIANA (THERIDIIDAE, ARANEAE) Mary E. A, Whitehouse Department of Zoology University of Canterbury Christchurch 1, New Zealand ABSTRACT The spider Argyrodes antipodiana (O.P. Cambridge) from New Zealand is a kleptoparasite whose primary host in nature is an orb weaving spider, Aranea pustulosa (Walckenaer). The kleptoparasite’s bias towards this host is stronger in the summer than in the winter. In the laboratory, Argyrodes was significantly better at obtaining food on the webs of Aranea pustulosa , than on the webs of Achaearanea sp., and Badumna longinquus (L. Koch). Factors that may be responsible for host preferences and for variation in efficiency on different types of webs are discussed. INTRODUCTION Argyrodes , a large cosmopolitan genus of theridiid spiders, is notorious for its kleptoparasitic species (Kullmann 1959; Vollrath 1979, 1979a, b; Smith Trail 1980; Rypstra 1981; Wise 1982; Larcher and Wise 1985; Whitehouse 1986). Instead of building prey-capture webs as do most theridiid spiders, Argyrodes roams through the periphery of other spiders’ webs gleaning trapped insects from the silk, pilfering wrapped food bundles directly from the resident spider (host), and sometimes attacking and eating the host. Vollrath (1984) suggested that Argyrodes can be loosely classified into two groups: Generalists and Specialists. Generalists invade a wide variety of web-types but use only a few techniques to obtain food; while specialists invade the webs of only a few species and use several techniques to obtain food. To be a specialist, Argyrodes often needs to respond opportunistically to the host’s movements, as it frequently feeds with the host or steals food bundles the host is guarding. Thus a specialist’s ability to choose the appropriate host is very important. Argyrodes antipodiana (O.P. Cambridge) (hereafter referred to as Argyrodes) is a kleptoparasitic spider from New Zealand. The behavioral repertoire of this spider is that of a specialist (Whitehouse 1986). Casual field observations made during the course of this earlier study suggested that Argyrodes tends to be highly restricted in its host-choice. The aims of this paper are to present more precise information on the host- choice of Argyrodes and to investigate possible reasons for restricted host-choice by this species. 350 THE JOURNAL OF ARACHNOLOGY METHODS Field surveys of hosts of Argyrodes. — Two surveys, one in early winter, May (approx, average daily temperature range = 4-15°C) 1984, the other in summer, January (approx, average daily temperature range = 15-25° C) 1985, were undertaken at Te Aroha (North Island, New Zealand: 37.32° S; 175.43° E) by examining all the webs in the sample area (ca. 50 m2), collecting any Argyrodes found, and recording the types of web on which they were found. A casual survey was conducted in late winter/ early spring, August (approx, average daily temperature range = 5-15°C) 1985, where the author walked over the sample area and noted the sex and maturity of the population. Laboratory analysis. — Spiders were maintained and tested in transparent plastic cages in a laboratory with controlled light (12:12, L:D) and temperature (20° C- 25° C) (for details see Jackson 1974). Locomotion on webs : Spider webs can be divided into three categories: cribellate webs, which are sticky because they are covered by very fine strands of silk; non-cribellate sticky webs, the glue of which consists of droplets of a sticky fluid; or non-cribellate non-sticky webs which have no glue (see Foelix 1982). Argyrodes was placed onto the three types of webs and its locomotion observed. Mortality on webs : Adult and sub-adult Argyrodes were housed upon the established webs of host species Badumna longinquus (L. Koch) (Amaurobiidae), Achaearanea sp. (Theridiidae) and Aranea pustulosa (Walckenaer) (Araneidae) (hereafter referred to as “ Badumna ”, 44 Achaearanea ”, and 44 Aranea ” respectively), until the Argyrodes were eaten, they died of natural causes, or the time period for the experiment was completed (the experiment ran for 27 days). Each host was used once only, except for one Badumna which was used twice. Badumna built cribellate sticky space webs, Achaearanea built non-cribellate sticky space webs, while Aranea built non-cribellate sticky orb webs. The spiders were fed every 1-4 days. I recorded the length of time each Argyrodes survived on hosts’ webs, and the number of Argyrodes that were killed by the hosts. Survival, measured as spider-days of exposure (the number of days Argyrodes were exposed to the host) was compared among host species using survival rate analysis (Johnson 1979; Harris et al. in prep.). Comparison of the capture efficiency of Argyrodes on the webs of three host species'. Host species Aranea , Achaearanea , and Badumna of a similar size (ca. 7 mm) were collected and housed in cages suitable for their web-type. The hosts were given ca. 10 days to establish a web before a subadult (i.e., a spider one molt before maturity) Or adult Argyrodes (body length: ca. 3 mm) was introduced to the cage. At 1-4 day intervals a test was started by dropping 10-20 Drosophila melanogaster (Meigen) (fruit flies) onto the host’s web, then 30 min later dropping another 10 fruit flies onto the web (a variable time scale was used to avoid host satiation as satiated hosts are less likely to construct webs). I recorded whether or not Argyrodes obtained food during a 2 h test period. If the host’s web did not retain five or more flies, the results were discarded as at this level of prey availability I deemed it too difficult for Argyrodes to obtain food. Each test was assumed to be independent of each other as the Argyrodes were responding to new conditions. For instance, the time between tests had enabled the host to reconstruct its web and the distribution of restrained flies on the web varied greatly between tests. WHITEHOUSE— HOST CHOICE BY ARGYRODES 351 RESULTS Hosts of Argyrodes in nature. — Only juvenile Argyrodes were discovered during winter. Of the 133 found, 59% were associated with araneid webs (Table 1). Besides being on or near the araneid orbs which are the food capture webs of the host, many Argyrodes were found in the eggsac webs of Aranea crassa (Walck.) while the maternal spider was standing on the eggsac. Eggsac webs are non-sticky arrays of silk (ca. 5x5x9 cm) which surround the eggsac. Up to 15 juvenile Argyrodes (body length: 1.0-2. 5 mm) were found motionless in a single eggsac web. Of 95 Argyrodes (11 males, 12 females, and 72 juveniles) found during summer, 85% were associated with Aranea webs (Table 2). All adults were found on orb webs. The casual survey ( n = ca. 50 spiders) conducted to reveal population structure of Argyrodes in early spring revealed the presence of three adults (2 males, 1 female), numerous sub-adult males (ca. 20), and juveniles (ca. 30). Locomotion on webs. — Argyrodes stuck to the cribellate webs of Badumna. After landing on the web, Argyrodes “froze”, then carefully tried to remove any legs stuck to the silk. If successful, the spider proceded to clean the freed leg by moving the tarsi through its chelicerae. Often, however, Argyrodes had great difficulty in freeing legs and remained motionless on the web, for several minutes at a time, in a posture not normally associated with resting. If the spider was unable to free itself completely after ca. 10 min, I removed it manually and returned it to its own web. In contrast, Argyrodes was seen to walk through large glue droplets on non-cribellate sticky webs of Aranea without any apparent difficulty. Argyrodes also had no evident difficulty moving on the non-cribellate sticky webs of Achaearanea. Mortality on webs. — Argyrodes varied greatly in its ability to survive on the host’s web. Of the 6 Argyrodes placed on webs of Badumna , 5 were killed in 42 spider-days; of the 5 Argyrodes placed on webs of Achaearanea , 3 were killed in 51 spider-days; and 6 Argyrodes placed on webs of Aranea , none were killed in 81 spider-days although one died of natural causes (that is, it was found dead rather than eaten). Argyrodes survived significantly better on webs of Aranea compared with webs of Badumna (Z = 2.10, P < 0.05), and survival on webs of Achaearanea was intermediate to, and not significantly different from survival on webs of either of the other species ( Aranea versus Achaearanea : Z = 1.68 P < 0.1; Achaearanea versus Badumna : Z = 0.85, P < 0.1). Capture efficiencies. — Argyrodes varied significantly in its ability to capture food on the three types of host webs, being successful in capturing food in webs of Aranea in 81% of the trials (n = 22), successful in 44% of the trials on webs of Achaearanea (n = 18), and successful in only 6% of the trials on webs of Badumna (n = 17) (P < 0.001, = 21.42, 2 df). Foraging behavior. — Argyrodes obtained food by either feeding with the host, stealing the host’s food bundles, or capturing Drosophila caught on the host’s web. For all these methods of food capture, Argyrodes proceded through the following six steps: (1) Argyrodes stands in its cryptic posture, not responding to food; (2) Argyrodes stands in its alert posture; (3) Argyrodes moves on web but not apparently towards a food item; (4) Argyrodes moves towards a food item; (5) Argyrodes touches the food; (6) Argyrodes feeds. The “cryptic” and “alert” 352 THE JOURNAL OF ARACHMOLOGY Table 1. — Webs occupied during winter by juvenile Argyrodes (n = 133) expressed as percentages of the total number of Argyrodes found. Description of Condition of Position of Percentage of Host host’s web host’s web Argyrodes ’ web Argyrodes found Argyrodes pustulosa Non-cribellate Maintained & Attached to 47.4% (Araneidae) sticky orb used by host host’s web Aranea crassa Eggcase Maintained & Attached to 12% (Araneidae) lattice used by host host’s web Leucauge Non-cribellate Maintained & Attached to 0.8% dromedaria horizontal orb used by host host’s web (Araneidae) surrounded by a maze of threads Cyclosa Non-cribellate Maintained & Attached to 0.8% trilobata vertical orb used by host host’s web (Araneidae) Argyrodes with stabilimentum In isolation 7.5% antipodiana (Theridiidae) Isolated but 6.0% behind old, unidentified silk (13.5%) Achaearanea sp. Non-cribellate Maintained & Attached to 6.0 % (Theridiidae) sticky space web used by host host’s web Cambridgea sp. Non-cribellate Maintained & Attached to 6.6% (Agelenidae) non-sticky large sheet web (<100 cm2) used by host host’s web Stiphidion sp. Non-cribellate Maintained & Attached to 1.5% (Agelenidae) non-sticky small sheet web (< 1 0 cm2) used by host host’s web Phoicus sp. Non-cribellate Maintained & Attached to 0.8% (Pholcidae) non-sticky space web used by host host’s web Badumna Cribellate Maintained & Attached to 0.8% longinquus (Amaurobiidae) space web used by host host’s web Maintained & Isolated but 6.0% used by host behind host’s web In disrepair Attached to 3.8% & abandoned by host host’s web (10.6%) postures are described elsewhere (Whitehouse 1986). The closest step towards feeding reached by the Argyrodes during the observation period was recorded. In nearly all the tests on webs of Aranea , Argyrodes reached step 6 (Fig. 1). Argyrodes on the webs of Achaearanea either stopped at step 3 or continued until it obtained food (step 6); it rarely failed to obtain food once it had located it (Fig. 1). Argyrodes often made several attempts to obtain food on the webs of both Achaearanea and Aranea before succeeding. Argyrodes on the web of Badumna rarely passed step 3. WHITEHOUSE— HOST CHOICE BY ARGYRODES 353 Table 2. — Webs occupied during summer by Argyrodes ( n = 95: 23 adults, 72 juveniles) expressed as percentages of the total number of Argyrodes found. Adults were only found on Aranea pustulosa webs. Description of Condition of Position of Percentage of Host host’s web host’s web Argyrodes ’ web Argyrodes found Aranea pustulosa Non-cribellate Maintained & Attached to 82.1% (Araneidae) sticky orb used by host host’s web In disrepair Attached to 3.2% & abandoned by host host’s web (85.3%) Aranea crassa Eggcase Maintained & Attached to 12% (Araneidae) lattice used by host host’s web Cambridgea sp. Non-cribellate Maintained & Attached to 3.2% (Agelenidae) non-sticky large sheet web (<100 cm2) used by host host’s web Stiphidion sp. Non-cribellate Maintained & Attached to 1.0% (Stiphidiinae) non-sticky large sheet web (<10 cm2) used by host host’s web In isolation 5.3% Argyrodes antipodiana Isolated but 4.2% (Theridiidae) behind old, unidentified silk (9.5%) Unidentified Unidentified In disrepair No web present 1.0% & abandoned by host DISCUSSION Field surveys of hosts of Argyrodes. — Argyrodes were found to mainly kleptoparasitize the webs of a single host species, Aranea. This characteristic supports the conclusion gained from its wide range of foraging behaviors (Whitehouse 1986) that Argyrodes antipodiana is a specialist kleptoparasite. The field surveys also reveal that the population structure of Argyrodes appears to be seasonal. Argyrodes overwinter as juveniles, mature in spring, and reproduce in summer. More work is needed to determine if one generation survives for the whole year or if there are two generations, a short one which survives only through summer and a longer one which overwinters. Evidently Argyrodes was more restricted to webs of Aranea during the summer than during winter. This may be linked to the seasonal variation in population structure. Adult Argyrodes , which only exploited Aranea , were abundant in summer, scarce in spring, and absent in winter. The feeding and mortality experiments showed that Argyrodes was significantly better at obtaining food and surviving on the webs of Aranea than on any other webs. Thus adults which must reproduce within a short period of time (probably ca. one month) in summer, are apparently limited to the webs of Aranea from which they can obtain food. Juvenile Argyrodes are able to survive for a long time in the laboratory (three months) without feeding (unpubl. data). While they are overwintering they need a web for shelter only, and so would not be restricted to the webs of Aranea. In 354 THE JOURNAL OF ARACHNOLOGY Fig. 1. — Distribution of the final behavioral step (as defined in foraging behavior section of results) that Argyrodes reached during a feeding bout: a, Argyrodes on the webs of Aranecr, b, Argyrodes on the webs of Achaearanea', c, Argyrodes on the webs of Badumna. Behavioral steps spring, when they need food to grow and mature, they apparently move to the webs of Aranea. Steps towards obtaining food. — The sequence of behaviors leading towards food acquisition was arrested for many spiders at step 3 (Fig. 1). In particular, nearly all spiders on the webs of Badumna and half on the webs of Achaearanea stopped at this point. Spiders that proceded past step 3 (moving on the web) usually persevered and continued to approach food items until they managed to obtain one. This observation suggests that Argyrodes may be better at obtaining food on webs of Aranea because it is unable to interpret vibrations on the webs of Badumna and, to some extent, Achaearanea. That is, Argyrodes appears capable of sensing vibrations upon the webs of Badumna and Achaearanea (in that it responds to the vibrations by moving), but is apparently unable to determine the direction from which the vibrations are coming. A complicating factor on the web of Badumna , however, is that Argyrodes is unable to walk on these webs. Nevertheless, Argyrodes uses its own web as a scaffolding to approach food on a hosts’ web (Whitehouse 1986) and so could conceivably use this to approach food on the web of Badumna and thus avoid, to a large extent, walking on the web of this host. Host preference. — The ability of Argyrodes to inhabit the webs of Badumna (cribellate, sticky, space web), Achaearanea (non-cribellate, sticky vertical orb WHITEHOUSE— HOST CHOICE BY ARGYRODES 355 web) was examined by looking at three parameters: the abilities to move, survive, and feed on the host’s web. Argyrodes was able to walk on webs of both Aranea and Achaearanea, but they became ensnared by the cribellate glue on webs of Badumna. In both ability to survive and feed, Argyrodes performed best on webs of Aranea , worst on webs of Badumna , and intermediately on webs of Achaearanea. Thus these parameters are probably major factors limiting adult Argyrodes to the webs of Aranea in the field. It is interesting, however, that other non-cribellate sticky orb webs, such as those of Cyclosa trilobata (Urqu.) which, common along with Aranea in the habitat of Argyrodes antipodiana , were not exploited. Thus not all orb webs and their residents fulfill the criteria upon which Argyrodes antipodiana bases its choice of hosts. ACKNOWLEDGEMENTS I would like to thank I. Plant for his support and constructive criticism of the manuscript, L. de Groot for constructive comments on the manuscript and its presentation, and D. Court for identifying some of the host species. I am grateful to I. Mclean for his critical comments and help with statistical analysis, and R. R. Jackson for his enthusiasm, criticism, and help at all stages of this work. REFERENCES Foelix, R. F. 1982. Biology of Spiders. Harvard Univ. Press, Cambridge, Massachusetts. Jackson, R. R. 1974. Rearing methods for spiders. J. Arachnol., 2:53-56. Johnson, D. H. 1979. Estimating nest success: the Mayfield method and an alternative. Auk, 96:651- 661. Kullmann, E. 1959. Beobachtungen und Betrachtungen zum Verhalten der Theridiide Conopistha argyrodes Walckenaer (Araneae). Mitt. Zool. Mus. Berlin, 35:275-292. Larcher, S. F. and D. A. Wise. 1985. Experimental studies of the interactions between a web-invading spider and two host species. J. Arachnol, 13:43-59. Rypstra, A. L. 1981. The effect of kleptoparasitism on prey consumption and web relocation in a Peruvian population of the spider Nephila otavipes. Oikos, 37:179-182. Smith Trail, D. S. 1980. Predation by Argyrodes (Theridiidae) on solitary and communal spiders. Psyche, 87:349-355. Vollrath, F. 1976. Konkurrenzvermeidung bei tropischen kleptoparasitischen haubennetzspinnen der Gattung Argyrodes (Arachnida: Araneae: Theridiidae). Entomol. Germ., 3:104-108. Vollrath, F. 1979a. Vibrations: their signal function for a spider kleptoparasite. Science, 205:1149- 1151. Vollrath, F. 1979b. Behavior of the kleptoparasitic spider Argyrodes elevatus (Araneae, Theridiidae). Anim. Behav., 27:515-521. Vollrath, F. 1984. Kleptobiotic interactions in invertebrates. Pp. 61-94, In Producers and Scroungers: Strategies of Exploitation and Parasitism. (C. Barnard, ed.) Croom Helm, London and Sydney. Wise, D. H. 1982. Predation by a commensal spider, Argyrodes trigonum , upon its host: An experimental study. J. Arachnol., 10:111-116. Whitehouse, M. E. A. 1986. The foraging behaviors of Argyrodes antipodiana (Theridiidae), a kleptoparasitic spider from New Zealand. New Zealand J. Zool, 13:151-168. Manuscript received December 1986, revised May 1988. Goloboff, P. A. 1988. Xenonemesia, un nuevo genero de Nemesiidae (Araneae, Mygalomorphae). J. Arachnol., 16:357-363. XENONEMESIA, UN NUEVO GENERO DE NEMESIIDAE (ARANEAE, MYGALOMORPHAE) Pablo A. Goloboff Museo Argentine de Ciencias N at u rales “Bernardino Rivadavia” Avda. Angel Gallardo 470 1405 Buenos Aires, Argentina ABSTRACT Xenonemesia platense, a new genus and species of nemesiid spider from Argentina and Uruguay, is described and figured. The new genus is characterized by having a wide sternum, slightly raised tarsal organ, slight scopula, apical article of posterior spinnerets domed, and by the absence of serrula, male tibial apophyses, keels on the bulb, and third claw. EXTRACTO Se describe e ilustra a Xenonemesia platense , un nuevo genero y especie de Argentina y Uruguay. El nuevo genero se caracteriza por tener esternon ancho, organo tarsal ligeramente elevado, escopula rala, artejo apical de las hileras posteriores hemisferico, y por la ausencia de serrula, apofisis tibiales en el macho, carenas en el bulbo y tercer una. INTRODUCCION Todos los generos de Nemesiidae descriptos de Sudamerica, excepto Spelocteniza Gertsch (Nemesiidae cavernicola incertae sedis) y Acanlhogonatus Karsch (Anaminae), pertenecen a las subfamilias Pycnothelinae y Diplothelopsi- nae (Raven 1985). El descubrimiento de una nueva especie de Argentina y Uruguay que no pertenence a ninguno de estos grupos, ni puede tampoco ser ubicada en ninguno de los grupos de Nemesiidae reconocidos por Raven (1985) lleva a describir un nuevo genero. MATERIALES Y METODOS Las medidas estan dadas en milimetros. La notacion de la denticion de unas, quetotaxia y tricobotriotaxia se hace segun Goloboff y Platnick (1987). Otras abreviaturas utilizadas son las usuales, y se pueden encontrar en Galiano (1970). Siguiendo a Coyle (1974) se denomina cerda ensiforme a aquella que tiene su extreme romo, y atenuada a aquella que se adelgaza gradualmente. El material estudiado esta depositado en las siguientes instituciones: Facultad de Ciencias Exactas y Naturales de la Universidad de Buenos Aires (FCEN), a cargo del Dr. Juan C. Giacchi; Museo Argentine de Ciencias Naturales “Bernardino Rivadavia” (MACN), a cargo del Dr. Emilio A. Maury; American Museum of Natural History (AMNH), a cargo del Dr. Norman I. Platnick. 358 THE JOURNAL OF ARACHNOLOGY Xenonemesia , nuevo genero Especie tipo. — Xenonemesia platense, n. sp. Etimologia. — El nombre generico, derivado de Nemesia (genero tipo de la familia) y la palabra griega xeno (extrano), se refiere a que el nuevo genero no parece estar emparentado cercanamente con ninguno de los de la familia Nemesiidae previamente descriptos. Es de genero femenino. Diagnosis. — Se diferencia de los demas generos conocidos de la familia por presentar simultaneamente ester non ancho, cymbium sin cerdas engrosadas, bulbo sin carenas, tibia I del macho sin apofisis, escopula tarsal poco deesa en tarsos anteriores y ausente en los posteriores, aha tarsal inferior ausente y artejo apical de las hileras posteriores hemisferico. Los demas generos de Nemesiidae de Argentina o Uruguay se difereecian facilmente de Xenonemesia por tener escopula tarsal y metatarsal densa en patas anteriores (y casi siempre, escopula en tarsos III). Descripcion. — Ver descripcion de la especie tipo. Xenonemesia platense , nueva especie Figs. 1-14 Tipos. — Holotypus hem bra de Argentina, provincia de Buenos Aires: General Pacheco, 27 IX 1980 (P Goloboff), MACN 8603. Los siguientes paratypi: Argentina, provincia de Entre Rios: Ao. Gualeyan, 5-6 II 1983 (R Goloboff), 3 machos, 4 hembras (MACN 8608, 8609); Parque Nacional El Palmar, 18 IV 1981 (P. Goloboff, A. Zanetic), 1 hembra (AMNH). Etimologia.— El nombre espedfico se refiere a la distribucion de la especie, que comprende Argentina y Uruguay, parses que hasta principios del siglo pasado formaron parte del virreinato del Rio de la Plata. Diagnosis. — Se reconoce facilmente por los caracteres genericos y por su colorido, con tres fajas claras en el cefalotorax y el abdomen sin “chevron” (Figs. 1,5). Descripcion de la hembra holotypus. — Largo total, 15.85. Cefalotorax (Fig. 1) de largo 5.66, ancho 4.53; RC convexa; RT mas baja y declive hacia atras; ancho 0.80 del largo. RC de ancho 0.77 de su largo; largo 0.70 del largo del cefalotorax; ancho 0.86 del ancho de la RT. Fovea ligeramente procurva; ocupa 0.15 del ancho de la RT. Tuberculo ocular elevado, ocupa 0.30 del ancho de la RC. OMA ligeramente mayores que los OMP; OLA ligeramente mayores que los OLR En el tuberculo ocular, 3 cerdas gruesas y 8 mas pequenas por delante de los OMA; 10 cerdas por detras de los OMA. En el clipeo, 7 cerdas dirigidas hacia adelante. Por detras del tuberculo ocular, una fila media de 3 cerdas gruesas y largas y dos filas laterales de unas 5 cerdas mas pequenas. Queliceros con rastrillo debit, formado por cerdas gruesas atenuadas; margee externo inerme, interno con 7 dientes que decrecen en tamano hacia el apice; parte basal del canal con 9 denticulos; cara anterior sin pelos clavados; gancho no aserrado y sin carenas. Coxas de los palpos con 12 a 17 espinulas, sin serrula. Labio inerme, de largo 0.48 del ancho. Esternon (Fig. 2) con su margen ligeramente rebordeado, de ancho 0.95 del largo, revestido de cerdas duras (sobre todo en el margen); sigillas pequenas, ovales, submarginales. GOLOBOFF— UN NUEVO GENERO DE NEMESIIDAE 359 Figs. 1-12. — Xenonemesia platense: 1-4, Hembra; 5-12, Macho; 1, cefalotorax y abdomen; 2, 6, esternon, labio y coxas de los palpos; 3, espermatecas, ventral; 4, hileras, ventrolateral; 5, cefalotorax; 7, quelicero derecho, cara anterior; 8, palpo izquierdo, prolateral; 9-10, bulbo derecho, caras opuestas; 11, detalle del margen del esternon (con sigilla III); 12, pata I derecha, prolateral; (1-2, Holotypus 8603 MACN; 3, 8602 MACN; 4, 8601 MACN; 5-6, 11, 8604 MACN; 7-10, 12, Paratypus 8608 MACN.). Escala = 0.5 mm. 360 THE JOURNAL OF ARACHNOLOGY Figs. 13-14. — Xenonemesia platense, hembra AMNH: 13, tricobotria del tarso IV; 14, organo tarsal del tarso IV. Pata I distinta de la II, mas gruesa, inerme, con sus unas mas cortas y gruesas; metatarso conico, adelgazado hacia del extreme distal. Escopula muy rala en tarso I, mas rala aun en tarso II y tercio apical de metatarso I, ausente en tarsos III y IV y metatarsos II-IV. Todas las patas sin peines metatarsales. Organo tarsal ligeramente elevado y convexo (Fig. 14). Cuticula de las patas lisa (Figs. 13, 14). Tarsos I-IV integros. Medidas de las patas: Femur Patella Tibia Metatarso Tarso Total I 3.53 2.53 2.53 1.50 1.03 11.12 II 3.00 2.00 1.93 1.37 1.20 9.50 III 2.60 1.40 1.67 1.90 1.33 8.90 IV 3.63 2.13 2.73 3.06 1.47 13.02 Palpo 2.20 1.27 1.43 — 1.13 6.03 Unas tarsales superiores: pata I, ambas unas, T-T ext., t int.; II, ambas unas con dos filas de T-T-T; III y IV, ambas unas con T-T-T-T ext., T-T-T int.; palpo, T-T-T en promargen. Una inferior ausente en todas las patas. Tricobotrias: Tibias I-IV con dos filas de 5 a 6 en la 1:2 B a 3:4 B; tibia del palpo, fila ant. 6 (1:1), post. 7 (1:1). Metatarsos con una fila diagonal y un grupo de 3 o 4 en el apice; I, 12 (2:3 A); II, 11 (2:3 A); III, 10 (3:4 A); IV, 12 (3:4 A). Tarsos con una fila en zig-zag en los 3:4 A; tarso I, 10; II-IV, 12; palpo, 9 (2:3 A). Botria ligeramente corrugados (Fig. 13). Quetotaxia: Femures inermes. Patella III, 1-1-1 P, 1 r; IV, 1 r; I, II y palpo, inermes. Tibia II, 1 p sup (1:3 a), 1 v ant a; III, 1-1 P (2:3 B), 1 D A, 3 V A; IV, 1-1/0-1 R, 0-2/ 1-2 V A; palpo, 2 V ANT A; I, inerme. Metatarso II, 1-0-1 D POST, 1-1 V POST (2:3 A); III, 1-1-1 P, 1-1 D ANT (1:2 A), 1-0-1 D POST, 1 R A, 2-2 V (1:2 A); IV, 2-1-1-2-2/2-2-2-2 V ANT, 1-1-2 V 6 2 V A, 1-1-1/ 1-1 R SUP, 1-1 D ANT (1:3 A); I, inerme. Todos los tarsos inermes. Hileras (Fig. 4): anteriores de largo 0.63, con 5 fusulas apicales; posteriores con sus fusulas regularmente distribuidas y de tamano uniforme; artejo basal de largo 0.67, con 25 fusulas (2:3 A); medio, 0.43, con 30 fusulas; apical, 0.27, con 25 fusulas, hemisferico. Espermatecas como en la Fig. 3. Todo el cuerpo revestido con cerdas delgadas, atenuadas, y pelos en forma de baston, muy pequenos y de color claro, acostados. Borde anterior del abdomen GOLOBOFF— UN NUEVO GENERO DE NEMESIIDAE 361 con cerdas engrosadas parecidas a las de Migas vellardi (Goloboff y Platnick, 1987, fig. 7), pero dirigidas hacia afuera y mas largas y rectas. Cefalotorax marron rojizo oscuro, con 3 bandas longitudinales mas claras bien nitidas. Patas de color marron, con manchas oscuras en apice de femures, tibias y metatarsos y en patellas. Abdomen oscuro, sin “chevron”, con manchitas blancas mas numerosas a los lados del area cardiaca (Fig. 1). Vientre de color claro con manchas oscuras. Descripcion del macho paratypus 8608 MACN. — Largo total 10.20. Cefalotorax (Fig. 5) de largo 4.26, ancho 3.26, menos elevado y convexo que en la hembra; ancho 0.77 del largo. RC de ancho 0.60 de su largo; largo 0.42 del largo del cefalotorax; ancho 0.51 del ancho de la RT. La fovea ocupa 0.11 del ancho de la RT. Los ojos ocupan 0.38 del ancho de la RC. Por delante de los OMA, 7 cerdas; 5 en el clipeo; 5 por detras de los OMA; una fila longitudinal de 5 por detras del tuberculo ocular y dos filas de cerdas (mas pequehas) a los costados. Margen del cefalotorax con cerdas fuertes y gruesas. Queliceros con denticion y rastrillo semejantes a los de la hembra, con tumescencia interqueliceral bien evidente, pequena, cubierta con pocas cerdas cortas (Fig. 7). Coxas de los palpos con 12 a 13 espinulas. Labio inerme, de largo 0.40 del ancho. Esternon (Fig. 6) de ancho 0.93 del largo, revestido con cerdas mas gruesas que en la hembra (Fig. 11). Pata I (Fig. 12) sin apofisis ni carenas de ningun tipo; metatarso recto, cilindrico. Escopula muy rala (mas aun que en la hembra) en tarsos I y II, ausente en tarsos III y IV y metatarsos I-IV. Tarsos I-IV integros. Medidas de las patas: Femur Patella Tibia Metatarso Tarso Total I 3.10 1.93 2.50 1.80 1.33 10.66 II 2.73 1.60 2.00 1.80 1.37 9.50 III 2.56 1.33 1.80 2.20 1.47 9.36 IV 3.26 1.73 2.73 3.33 1.57 12.62 Palpo 1.83 0.90 1.73 — 0.73 5.19 Unas tarsales superiores con doble fila de dientes: pata I, ambas unas 5 int., 6 ext.; II, ambas unas con dos filas de 7; III, una ant. 8 int., 9 ext., post. 10 int., 7 ext.; IV, una ant. con dos filas de 9, post. 8 int., 7 ext. Tricobotrias: Tibias: I, II y IV, 5 6 6 en cada fila (ocupando 2:3 a 3:4 B en la fila ant., mas extendidas en la posterior); III, ant. 4 (1:2 B), post. 4 (2:3 B); palpo, ambas filas 5 (1:1). Metatarsos: I, II y III, 8 (2:3 A); IV, 11 (2:3 A). Tarsos I-IV, 8 a 10 (2:3 A); palpo, 9 (1:3 M). Quetotaxia: femur I, Ll-1-1-2 D; II, LI D ANT (1:3 A), Ll-Ll D, 1-1-1 D POST; III, 1-1 D ANT (1:2 M), 1-1-1 D (1:2 B), 1-1-1 D POST (1:2 A); IV, 1-1-1 D (2:3 B); palpo, LL1-1-2 D. Patella I, 1 p sup (1:3 a), 1 V A; II, L 1/1 P SUP; III, 1-1-2 P, 1 R; IV, 1 R; palpo, 4/5 d a. Tibia I, 3 P M, 2-L0-2-1 V, 2 V ANT A; II, 1-1 P SUP, Ll-Ll V (alternadas), 3 V A; III, LI P, 2-1 D, LI R, 2-2-3 V; IV, 1-1 P, 1 D B, LI R, 2-2-3 V; palpo, 1 P A, 1 R A. Metatarso I, 1 V B, 3-2 V A; II, LI P SUP, 2-2-3 V; III, 1-1-1 P, 1-1-2 D, 2-2 V, verticilo A de 5; IV, 1-1-1 P, 1-1-2 D, 2-L2 V, verticilo A de 5. Tarsos I-IV inermes. Palpo como en la Fig. 8; cymbium inerme, sin cerdas engrosadas; tibia con su excavacion ventral poco profunda. Bulbo (Figs. 9-10) sin carenas, con el embolo delgado. Area epigastrica con c. 35 glandulas epiandricas. 362 THE JOURNAL OF ARACHNOLOGY Todo e! cuerpo revestido con cerdas mi.mei.osas, mas cortas y gruesas que en la hern bra. ensiformes, excepto en los artejos apicales de patas y paipos, que tienen cerdas atenuadas. Colorido semejante al de la hembra. Yariaciones. — No se observaron variaciones de importaecia en los caracteres mencionados, excepto en la forma de la fovea; algunos ejemplares presentae la fovea con sus bordes ligeramente recurvados (como el macho 8604 MACN, Fig. 5). En uno de los machos de General Pacheco hay solo 11 V, 1-1 P, 1 Y ANT en tibia I; en los demas machos examinados la quetotaxia es similar a la del paratypus 8608 MACN. Historia natural. — En el Parque National El Palmar Xenonemesia platense fue colectada en la barranca del rio Uruguay, bajo piedras, en un ambiente burned© y sombrio, junto con Grammostola sp., Homoeomma uruguayensis (Mello-Leitao, 1946), Stenoterommata argentinensis (Sehiapelli y Gerschman de Pikelin, 1958), Stenoterommata sp., Actinopus sp. e Idiops dams (Mello-Leitao, 1946). En el arroyo Gualeyan, en lugar llano, se la encontro en un monte mas abierto y xerofilo (donde abundan plantas como Opuniia y Aspidosperma ), en los monticuios de tierra al pie de arboles. Otras Mygalomorphae colectadas en este lugar fueron Grammostola sp. (distinta de la de P. N. El Palmar), Eupalaestrus campestratus (Simon, 1897), Stenoterommata sp., Actinopus sp. e Idiops dams, En General Pacheco, en un ambiente alter ado. Xenonemesia platense fue encontrada en los monticuios de tierra al costado de los caminos en un parque con suaves barrancas, donde no se encontraron otras Mygalomorphae. Los ejemplares fueron ha H ad os en cuevas poco profundas (meeos de 10 cm de profundidad), sin operculo, de recorrido bastante irregular, de 1 cm de ancho aproximadamente, con sus paredes cubiertas con muy poca seda; habitualmente cierran la cueva durante el dia amontonando tierra y seda en la entrada. Un ejemplar colectado con la ooteca estaba en una camara oval, de 20 o 25 mm de largo y 10 o 15 de ancho, cerrada, con el resto de la cueva tapado. Se observe que ejemplares en cautiverio pueden capturar sus presas sin tener cueva construida. Mientras comen la presa, depositae seda en el sustrato, con movimientos circulates similares a los de Theraphosidae (Eberhard 1967). Material examinado. — ARGENTINA: BUENOS AIRES; Buenos Aires, 1979 (D. Martinez), 1 hembra, 1 macho jov. (FCEN), General Pacheco, 28 X 1979 (P. Goloboff), 2 hembras (MACN 8601), 5 I 1980 (P. Goloboff), 3 hembras (MACN 8602), 27 IX 1980 (P. Goloboff), 2 machos, 1 hembra (MACN 8604), 18 X 1980 (P. Goloboff), 1 hembra (MACN 8605). ENTRE RIOS; Parque Nac. Ei Palmar, 12-16 II 1980 (P. Goloboff), 3 hembras, 2 machos jov. (MACN 8606), 18 IV 1981 (P. Goloboff, A. Zanetic), 1 macho jov. (MACN 8607), Arroyo Gualeyan, 5-6 II 1983 (P. Goloboff), 2 machos jov., 1 hembra jov. (MACN 8618), 13 X 1984 (P. Goloboff, C. Szumik), 2 hembras, 1 macho jov., -1 hembra jov. (MACN 8610), 27 IX 1987 (P. Goloboff, C. Szumik), 1 macho, 1 hembra, 2 jovs. (MACN 8619). URUGUAY: LAVALLEJA; Picada de Rodriguez, 8 VII 1957 (sin colector), 1 macho joven (MACN 8611), bajo piedra. COLONIA; San Juan, XII 1950 (R. Ringuelet), 1 macho jov. (MACN 8612). RELACIONES DE XENONEMESIA CON OTRAS NEMESIIDAE Las unicas sinapomorfias de ia familia Nemesiidae son las unas tarsales superiores anchas y con dientes biseriados y la una del palpo de la hembra con dientes en promargen; dichos caracteres sostieeen la inclusion de Xenonemesia en GOLOBOFF— UN NUEVO GENERO DE NEMESIIDAE 363 Nemesiidae, al menus mientras no se acepte la hipotesis alternativa de que son una sinapomorfia de Crassitarsae (Theraphosoidina mas Nemesiidae) y no solo de Nemesiidae (Raven 1985:28). Las interrelaciones de las Nemesiidae son un poco inciertas; el cladograma que Raven (1985, fig. 4) ha presentado tiene un gran numero de homoplasias, como ya fue destacado por su mismo autor (1985:46). El descubrimiento de Xenonemesia complica aun mas el panorama, porque este genero comparte algunos caracteres derivados con algunos grupos de Nemesiidae y otros caracteres con otros grupos. La una tarsal inferior falta solo en Diplothelopsinae y algunas Acanthogonatus y Pycnothelinae. El esternon es ancho en algunas Bemmerinae y Diplothelopsinae (no en todas, como cita Raven 1985:97). El artejo apical de las hileras laterales posteriores es hemisferico en Nemesiinae y algunas Ixamatinae, Pycnothelinae, Bemmerinae y Acanthogonatus. La serrula falta tambien en Nemesiinae y algunas Ixamatinae, Pycnothelinae y Anaminae. Dado que segun los caracteres mencionados la inclusion de Xenonemesia en cualquiera de estos grupos requiere un grado de homoplasia mas o menos similar, sus relaciones con las demas Nemesiidae permanecen inciertas por ahora. AGRADECIMIENTOS El Dr. Norman I. Platnick, con la gentileza habitual en el, tomo las fotomicrografias que ilustran este trabajo y, al igual que el Dr. Robert J. Raven (Queensland Museum, Australia) y Prof. Maria E. Galiano (MACN), hizo la lectura critica del manuscrito. LITERATURA CITADA Coyle, F. A. 1974. Systematics of the trapdoor spider genus Aliatypus (Araneae: Antrodiaetidae). Psyche, 81:431-500. Eberhard, W. G. 1967. Attack behavior of diguetid spiders and the origin of prey wrapping in spiders. Psyche, 74:173-181. Galiano, M. E. 1970. Revision del genero Tullgrenella Mello-Leitao, 1941 (Araneae, Salticidae). Physis (Buenos Aires), 29(79):323-355. Goloboff, P. A. y N. I. Platnick. 1987. A review of the Chilean spiders of the superfamily Migoidea (Araneae, Mygalomorphae). American Mus. Novitates, 2888:1-15. Raven, R. J. 1985. The spider infraorder Mygalomorphae (Araneae): cladistics and systematics. Bull. American Mus. Nat. Hist., 182(1): 1-180. Manuscript received January 1988, revised May 1988. Sissom, W. D. 1988. Typhlochactas mitchelli , a new species of eyeless, montane forest litter scorpion from northeastern Oaxaca, Mexico (Chactidae, Superstitioninae, Typhlochactini). J. Arachnol., 16:365-371. TYPHLOCHACTAS MITCHELLI, A NEW SPECIES OF EYELESS, MONTANE FOREST LITTER SCORPION FROM NORTHEASTERN OAXACA, MEXICO (CHACTIDAE, SUPERSTITIONINAE, TYPHLOCHACTINI) W. David Sissom Department of Biology Elon College Elon College, North Carolina 27244 USA ABSTRACT Typhlochactas mitchelli , new species, is described from Cerro Ocote, near Tenango, Oaxaca, Mexico. This is the second species of Typhlochactas discovered in montane forest litter. Based on its cheliceral dentition, T. mitchelli is most closely related to the other forest litter species, T. sylvestris Mitchell & Peck, also from Oaxaca. INTRODUCTION The first eyeless scorpion from montane forest litter was discovered along the east slopes of the outer range of the Sistema Montanoso Poblano Oaxaqueno near Valle Nacional, Oaxaca by Dr. Stewart B. Peck in May of 1971 (Mitchell and Peck 1977). This discovery was highly significant because it was the first species of the genus Typhlochactas (all of which are eyeless, depigmented scorpions) collected outside the cave environment. Typhlochactas now consists of four species: T. rhodesi Mitchell from La Cueva de la Mina in Tamaulipas; T reddelli Mitchell from La Cueva del Ojo de Agua de Tlilapan in Veracruz; T. sylvestris Mitchell and Peck from montane forest litter in Oaxaca; and T. cavicola Francke from La Cueva del Vandalismo in Tamaulipas (Mitchell 1968; Mitchell and Peck 1977; Francke 1986). A fifth species, 7. elliotti Mitchell from El Sotano de Yerbaniz in San Luis Potosi, has been transferred to a separate genus, Sotanochactas (Mitchell 1971; Francke 1986). It is the purpose here to describe another species of this remarkable genus, the second one from montane forest litter. It is most closely related to T. sylvestris, the other forest litter species, but differs from it in a number of significant features. The new species was collected on Cerro Ocote near Tenango, Oaxaca along the northeastern edge of the Sistema Montanoso Poblano Oaxaqueno. Typhlochactas mitchelli, new species Figs. 1-14 Type data. — Holotype male, paratype male, and subadult paratype female taken from Cerro Ocote, 5 mi S Tenango, Oaxaca, Mexico in April 1987 (A. 366 THE JOURNAL OF ARACHNOLOGY Fig. 1. — Dorsal view of holotype male of Typhlochactas mitchelli, new species. SISSOM— NEW TYPHLOCHACTAS FROM FOREST LITTER 367 Figs. 2-14. — External morphology of holotype male of Typhlochactas mitchelli, new species: 2, dorsal aspect of right chelicera; 3, ventral aspect of sternum, genital operculi, and pectines; 4, lateral aspect of metasoma and telson; 5, dorsal aspect of pedipalp femur; 6, dorsal aspect of pedipalp tibia; 7, external aspect of pedipalp tibia; 8, ventral aspect of pedipalp tibia; 9, dorsal aspect of pedipalp chela; 10, external aspect of pedipalp chela; 11, ventral aspect of pedipalp chela; 12, inner margin of pedipalp chela fixed finger, showing placement of trichobothria and dentition; 13, inner margin of pedipalp chela movable finger, showing dentition; 14, retrolateral aspect of tarsomere II of left leg IV. 368 THE JOURNAL OF ARACHNOLOGY Grubbs, A. Cressler, P. Smith). Holotype male, paratype male, and paratype subadult female deposited in the American Museum of Natural History, New York. Etymology. — The specific epithet is a patronym honoring Dr. Robert W. Mitchell of Texas Tech University, who inspired my initial interest in arachnids, for his contributions to Mexican scorpiology and biospeleology. Distribution. — Known only from the type locality. Diagnosis. — Adult males 8.49-8.99 mm long. Eyeless. Color pale yellow brown, except for posterior mesosomal segments and metasoma, which are light orange brown. Carapace, tergites, and metasoma sparsely to moderately finely granular; pedipalps more coarsely granular. Metasomal segment V slightly longer than carapace and about 1.85 times longer than wide. Cheliceral fixed finger with only three teeth; basal and medial teeth not combined into a compound tooth. Movable finger with four teeth: distal internal, distal external, medial, and basal. Pedipalps: trichobothrial pattern typical of genus (Mitchell and Peck 1977); chela relatively robust with palm length/ width ratio 1.71-1.73; chela fingers shorter than carapace; fixed finger of chela with four slightly oblique rows of granules on dentate margin, restricted to distal two-thirds of finger; movable finger with five such rows. Legs armed with prolateral pedal spurs; ventral aspect of tarsomere II with median row of minute spinules flanked by three to four pairs of relatively stout setae. Description. — Based on adult males; measurements of these two males are given in Table 1. Coloration: Prosoma and first six mesosomal segments pale yellow brown; mesosomal segment VII (tergite and sternite) slightly darker than preceding segments. Pectines whitish. Metasoma uniformly light orange brown. Telson pale yellow brown; aculeus orange brown. Chelicerae and legs pale yellow. Pedipalps uniformly pale yellow brown, slightly darker than body. Prosoma: Carapace (Fig. 1) subquadrate; length slightly greater than posterior width. Weakly sclerotized; surface sparsely, finely granular with a few small setae. Anterior margin weakly convex, with very subtle median projection. Median longitudinal furrow essentially obsolete. Median and lateral eyes absent; ocular tubercle absent. Sternum smooth, subquadrate: anterior margin gently convex, posterior margin concave, lateral margins diverging distally; small posteromedial depression present. Mesosoma: Tergites I-VII weakly sclerotized, acarinate; pre-tergites smooth; post-tergites moderately finely granular. Genital operculum (Fig. 3) subelliptical, completely divided longitudinally; genital papillae well developed. Pectines (Fig. 3): more or less unsclerotized, with three marginal lamellae, two middle lamellae, and five pectinal teeth. Proximal middle lamella much larger than second. Pectinal lamellae and basal portion of teeth moderately covered with fine whitish microchaetes; distal third of pectinal teeth with conspicuous, dense, peg sensillae. Sternites III-VII smooth, sparsely setose; stigmata small, circular. Metasoma (Fig. 4): Segments I-III wider than long; V 1.82-1.85 times longer than wide. Segments I-IV: Dorsolateral carinae on I-IV very faint, indicated by a few small distal granules; other carinae obsolete. Dorsal and lateral surfaces with moderately dense, fine granulation; ventral surfaces smooth to sparsely, finely granular; setation of first four segments sparse. Segment V: distinctly longer than carapace; dorsolateral carinae faint, granular throughout; other carinae obsolete. SISSOM— NEW TYPHLOCHACTAS FROM FOREST LITTER 369 Table 1. — Measurements in mm and pectinal tooth counts of the holotype and paratype males of Typhlochactas mitchelli, n. sp. Holotype Male Paratype Male Total length 8.99 8.49 Carapace length 1.17 1.14 Mesosoma length 2.66 2.43 Metasoma length 3.61 3.46 length/ width I 0.45/0.73 0.45/0.74 length/width II 0.52/0.68 0.50/0.67 length/width III 0.55/0.69 0.55/0.67 length/ width IV 0.80/0.71 0.72/0.65 length/ width V 1.29/0.71 1.24/0.67 Telson length 1.55 1.46 Vesicle length/ width/depth 1.17/0.73/0.59 1.05/0.70/0.55 Aculeus length 0.38 0.41 Pedipalp length 3.43 3.27 Femur length/ width 0.85/0.35 0.82/0.33 Tibia length/ width 0.98/0.39 0.90/0.38 Chela length/ width/ depth 1.60/0.52/0.55 1.55/0.51/0.55 Palm length 0.90 0.87 Fixed finger length 0.70 0.68 Movable finger length 0.92 0.90 Pectinal tooth count 5-5 5-5 Setation moderate, with most setae on ventral aspect. All surfaces with moderately dense, fine granulation. Dorsal surface with narrow median longitudinal furrow anteriorly and rounded, shallow depression posteriorly. Sum of metasomal I-V lengths 3.04-3.09 times greater than carapace length. Telson (Fig. 4): Vesicle flattened dorsally, moderately globose ventrally; telson as wide as first metasomal segment, wider than segments II-V. Lateral and ventral aspects of vesicle with moderately dense, fine granulation; about 20 pairs of setae. Aculeus very slender and strongly curved. Chelicerae: Fixed finger (Fig. 2) with only three individual teeth (distal, median, and basal). Movable finger (Fig. 2) with four teeth: distal internal tooth large, distinctly separated from others; distal external, medial, and basal teeth situated close together at midfinger; medial tooth smaller than either distal external or basal teeth. Distinct serrula present on ventrodistal two-thirds of movable finger. Dense array of long, thin setae present on medial and ventral surfaces of fixed finger; a few longer hairlike setae situated on ventral aspect of movable finger (proximal to serrula). Pedipalps: Femur (Fig. 5) with faint dorsoexternal carina present only on basal one-third; other carinae obsolete. All surfaces moderately granular. Orthobothrio- taxia C (Vachon 1974). Tibia (Figs. 6-8): carinae essentially obsolete, surfaces uniformly moderately granular. Orthobothriotaxia C (Vachon 1974); trichobo- thria db and dt petite; trichobothrium V2 located on external aspect (Fig. 7). Chela (Figs. 9-13): manus slightly swollen, with palm length/chela width ratio of 1.71-1.73; carinae essentially obsolete, but dorsal margin well supplied with coarser granules. All other surfaces moderately to densely granular. Fixed finger (Fig. 12) granular basally, with four slightly oblique rows of denticles limited to distal two-thirds of inner margin; basal row shortest; only three inner accessory granules paired with terminal denticle and enlarged granules of the two apical 370 THE JOURNAL OF ARACHNOLOGY rows. Movable finger (Fig. 13) granular basally, with five slightly oblique rows of denticles limited to distal two-thirds of inner margin; basal row short, about as long as apical row; four inner accessory granules paired with the terminal denticle and enlarged granules of two apicalmost rows. Movable finger as long as palm, but distinctly shorter than carapace or metasoma V; fixed finger length/ carapace length ratio of 0.60. Orthobothriotaxia C (Vachon 1974); trichobothria ib and it situated just basal to junction of fixed finger and manus (Figs. 11-12); trichobothria Db, Esb, Et4, Ets , and esb petite (Fig. 10). Legs: All segments moderately setose. No tibial spurs; only a single pedal spur located on prolateral aspect in arthrodial membrane between tarsomeres I and II (Fig. 14). Ventral aspect of tarsomere II (Fig. 14) with three to four pairs of setae flanking a median row of tiny spinules. Unguis moderately developed, weakly curved; dactyl (median claw) moderate. Variation. — There was no significant variation in the two male specimens. The subadult female was much paler in coloration, being more or less uniformly cream-colored. This specimen also retains vestigial rows of granules extending to near the base of the pedipalp chela fixed and movable fingers; therefore, it has five rows on the fixed finger and six rows on the movable finger. There are no enlarged basal granules or inner accessory granules on the fourth row on the fixed finger or on the fifth row of the movable finger. This information may indicate that reduction of the number of rows of granules as found in the adults occurs at the maturation molt. In addition, the cuticular surfaces were consistently less granular than in the males. The female also had a malformed right pectine with the two proximal pectinal teeth fused at the base. Comparisons. — Typhlochactas mitchelli differs from the other species of Typhlochactas by having only four rows of denticles on the chela fixed finger and only five on the movable finger. Further, these rows of denticles do not extend the full length of the fingers as in the other species. Typhlochactas mitchelli is most similar to T. sylvestris Mitchell and Peck, also from montane forest litter in Oaxaca, Mexico. Both of these species have only three individual teeth on the cheliceral fixed finger, a hypothesized synapomorphy (there are four teeth on the fixed finger in other Typhlochactas). There are three external teeth on the cheliceral movable finger in T. mitchelli and three or four in T. sylvestris (resulting from asymmetry in the holotype). However, the configuration of the teeth is quite different in the two species; the distal tooth in T. sylvestris is quite large compared to the others and more closely associated with the distal external tooth, rather than with the other external teeth as in T. mitchelli (Fig. 2). Typhlochactas mitchelli also differs from T. sylvestris in the more highly developed granulation of its tergites, metasoma, and pedipalps. The ventral aspect of tarsomere II bears a median spinule row in T. mitchelli , but not in T. sylvestris. There are also distinct differences in morphometries: in T. mitchelli , (1) the metasoma is proportionately longer, with the sum of metasomal I-V lengths/ carapace length 3.04-3.09 (not 2.51) and metasoma V length/ carapace length 1.09- 1.10 (not 0.99); (2) chela fixed finger length/ carapace length is 0.60 (not 0.72); and (3) chela palm length/chela width is 1.71-1.73 (not 1.48). Comments: Typhlochactas mitchelli and T. sylvestris are certainly the two smallest described scorpion species in the world. The total length of T. mitchelli ranges from 8.49-8.99 mm; that of the holotype (and only known specimen) of T. SISSOM— NEW TYPHLOCHACTAS FROM FOREST LITTER 371 sylvestris is reported to be 11.05 mm. However, examination of Mitchell and Peck’s (1977) table of measurements indicates a disproportionately large mesosomal length measurement, and it is apparent that the authors must have taken a single measurement of the mesosoma (rather than taking the sum of the lengths of the individual segments, as was done here). The intersegmental membranes stretch during preservation, and the degree of stretching will vary with the specimen. Without remeasuring the individual mesosomal tergites of T. sylvestris , it is difficult to say which of the two species is actually smaller; the carapace of T. mitchelli is shorter, but its metasoma and telson are larger. However, taking a single mesosomal measurement of the two adults of T. mitchelli results in total lengths of 9.46 and 9.90 mm, so T. mitchelli might be the smaller of the two and, therefore, the smallest known scorpion. Francke’s (1981) cladogram depicting the phylogeny of the Superstitioninae is not greatly modified by the addition of T. mitchelli. Typhlochactas mitchelli is added at the terminal branch as the sister species of T. sylvestris ; the synapomorphy justifying their relationship is the joint possession of three teeth on the cheliceral fixed finger. Reduction of the number of granular rows on the chela fingers and the unique configuration of the dorsal teeth of the cheliceral movable finger are autapomorphic characters for T. mitchelli. ACKNOWLEDGMENTS I am very grateful to Mr. James R. Reddell of the Texas Memorial Museum, Austin for giving me the opportunity to examine and describe this interesting species and for providing me with the appropriate geographical information. LITERATURE CITED Francke, O. F. 1981. Studies of the scorpion subfamilies Superstitioninae and Typhlochactinae, with description of a new genus (Scorpiones, Chactoidea). Assoc. Mexican Cave Stud. Bull., 8:51-61/ Texas Mem. Mus. Bull., 28:51-61. Francke, O. F. 1986. A new genus and a new species of troglobite scorpion from Mexico (Chactoidea, Superstitioninae, Typhlochactini). Texas Mem. Mus., Speleol. Monogr., 1:5-9. Mitchell, R. W. 1968. Typhlochactas , a new genus of eyeless cave scorpion from Mexico (Scorpionida, Chactidae). Ann. Speleol., 23:753-777. Mitchell, R. W. 1971. Typhlochactas elliotti , a new eyeless cave scorpion from Mexico (Scorpionida, Chactidae). Ann. Speleol., 26:135-148. Mitchell, R. W. and S. B. Peck. 1977. Typhlochactas sylvestris, a new eyeless scorpion from montane forest litter in Mexico (Scorpionida, Chactidae, Typhlochactinae). J. Arachnol., 5:159-168. Vachon, M. 1974. Etude des caracteres utilises pour classer les families et les genres de Scorpions (Arachnides). Bull. Mus. Nat. Hist, nat., Paris, 3rd ser., No. 140, Zool. 104:857-958. Manuscript received April 1988, revised May 1988. Welbourn, W. C. and O. P. Young. 1 988. Mites parasitic on spiders, with a description of a new species of Eutrombidium (Acari, Eutrombidiidae). J. Arachnol., 16:373-385. MITES PARASITIC ON SPIDERS, WITH A DESCRIPTION OF A NEW SPECIES OF EUTROMBIDIUM (ACARI, EUTROMBIDIIDAE) W, Calvin Welbourn Acarology Laboratory, Dept, of Entomology Ohio State University Columbus, Ohio 43210 USA and Orrey P, Young Southern Field Crop Insect Management Laboratory USDA, ARS, P. O. Box 346 Stonesville, Mississippi 38776 USA ABSTRACT A new species of Eutrombidium is described from larvae parasitizing 38 Ceraticelus emertoni (O. Pickard -Cambridge) (Araneae, Linyphiidae) and one Oxyopes salticus Hentz (Araneae, Oxyopidae) collected in Mississippi. Most host individuals (89%) were parasitized by only one larva, but as many as nine larvae were attached to one host. Adult and immature hosts of both sexes were parasitized. All larval mites were attached to the lateral molt sutures, mostly on the posterior prosoma. A review of the literature reveals 30 records of mite ectoparasitism of spiders among eight mite genera from five continents. Six additional records are reported herein. Two species listed as spider parasites, Allothrombium metae Boshell & Kerr (Acari, Trombidiidae) and Copriphis bristowi Finnegan (Acari, Laelapidae), are transferred to Clinotrombium and Ljunghia , respectively. INTRODUCTION Larvae of the cosmopolitan genus Eutrombidium Verdun (Acari, Eutrombidii- dae) parasitize a variety of Orthoptera (Welbourn 1983), whereas the active postlarval instars of at least one species, E. locustarum (Walsh), are predators of orthopteran eggs (Severin 1944). Of the 17 nominate species listed by Thor and Wilimann (1947), 14 were known from only the postlarval instars. Since then, three additional species have been described from orthopterans. Numerous species remain to be described worldwide. There are currently three available names for species of Eutrombidium in North America: E. locustarum , E. magnum (Ewing), and E . corticis (Ewing). Examination of the type of Ottonia trombidioides Banks indicates this species should be placed in Eutrombidium , E, trombidioides (Banks), new combination. Eutrombidium corticis should be placed in the Trombidiidae, possibly in the genus Allothrombium (Berlese). Two of the remaining three species, E. magnum and E. trombidioides , are known only from postlarval instars and need to be redescribed on the basis of reared larvae to determine their relationships with the 374 THE JOURNAL OF ARACHNOLOGY other named species. All North American larvae reported in the literature have been (mis-) identified as either E. trigonum (Hermann) or E. locustarum . Eutrombidium trigonum is an European species and its presence in North America has not been verified. Eutrombidium locustarum larvae have been reported from North American orthopterans representing more than 35 genera in four families (Welbourn 1983; Rees 1973; Huggans and Blickenstaff 1966). Examination of spiders collected in west central Mississippi revealed larvae of an undescribed species of Eutrombidium attached to two different spider species. The absence of previous reports of this mite genus parasitizing spiders and the inadequacy of larval characters used in earlier descriptions justifies our new generic diagnosis and description of the new species. A summary of the biology of this species and a survey and discussion of the general phenomenon of mite parasitism of spiders is also presented. TAXONOMY All measurements are in micrometers (/an) unless otherwise noted. Terminology generally follows Welbourn and Young (1987) and Robaux (1974). Genus Eutrombidium Verdun Eutrombidium Verdun 1909. Soc. Biol. 67:244; Type species: Trombidium trigonum Hermann 1804. Diagnosis. — Larva: Coxal field I with seta la nude; coxal fields I, II and III each with thickened and bifid seta, lb, 2b and 3b respectively; fri l r = 1-1-1; fnFe = 6-5-4; fnGe = 4-2-2; fnTi = 6-5-5; fsol = I (0-2-2-1), II (04-2-1), III (04-0-0); izeta - 2-1-0 or 2-0-0; famulus on tarsus leg I distal to omega ; palpal femur and genu each with a minute dorsal or lateral seta; one of three setae (in addition to palpal tibial claw) on palpal tibia spinelike or hypertrophied; palpal tibial daw bifurcate; scl hypertrophied. Deutonymph and Adult: Dorsal idiosomal setae set) form.; posterior idiosoma with pygosomal plate; palpal tibia with two rows of dorsal spines and one to four large ventral spines. Eutrombidium lockleii, new species Type data. — Holotype (AL-3280) and 55 paratypes ex Ceraticelus emertoni (O. Pickard- Cambridge) (Araneae, Linyphiidae) from Mississippi, Sunflower Co., 8 km SSW Indianola, in field dominated by coastal bermuda grass, collected by D- Vac suction method, 19 July 1984, T. C. Lockley. Two additional paratypes from same locality and date ex Oxyopes saiticus Hentz (Araneae, Oxyopidae). The holotype and four paratypes will be deposited in the United States National Museum, two paratypes each will be sent to the following institutions: Field Museum of Natural History, Chicago; Canadian National Collection, Ottawa; British Museum (Natural History), London; Museum National d’Histoire Naturelle, Paris; South Australian Museum, Adelaide; University of Michigan Museum of Zoology, Ann Arbor. The remaining paratypes will reside in the Acarology Laboratory, The Ohio State University, Columbus. WELBOURN & YOUNG — MITES PARASITIC ON SPIDERS 375 Diagnosis. — Larva with eyes and ocular sclerites incorporated into prodorsal sclerite; genu, legs I, II and III, each with at least one very long barbed seta; palpal tibial claw bifurcate distally and with basal knob; palpal tarsus with one very long barbed seta; lophotrix and scopa on tarsus leg III undeveloped; tarsus leg II without subterminal eupathid. Description. — Larva: Idiosoma (Figs. 2, 6). Holotype partially engorged. Due to distortion during mounting no size measurements were made; unmounted specimens ranged from 200 (unengorged) to 700 (engorged); eyes 2/2 incorporated into prodorsal sclerite, anterior eye smaller. Prodorsal sclerite and scutellum occupy most of the dorsal idiosoma in unengorged specimens, displacing dorsal idiosomal setae posteriorly and ventrally. Setal rows C and D each with three pairs of setae, rows E and F each with two pairs of setae; H and PS rows each with one pair of setae. Setae cl on scutellum; c2 and d\ each set on narrow sclerites. Idiosomal setae (Figs. I, 3) cl (59-71), d\ (50-63) longer than setae in rows E (13-29), F (13-20), and setae c2 (33-39), c3 (29-38), d2 (19-22), d3 (18-23); H and PS setae long, 38-51 and 56-67, respectively. Cupules and supracoxal seta (cl) absent. One pair of closely associated, branched intercoxal setae between coxae III; two pairs of preanal setae. Prodorsal Sclerite (Figs. 2, 5, 6): Punctate without striae, anterior margin convex, posterior margin slightly concave; PL > S > AL > AM; SB < PW; trichobothridial bases anterior to PL setal bases; trichobothria flagellate, with setules. Scutal measurements of holotype with mean, range and number of paratypes measured given within parentheses: AM 14 (14, 11-17, 19), AA - (61, 58-64, 10), AW - (85, 72-93, 5), AL 33 (33, 28-36, 21), PL 70 (68, 63-72, 22), AP 35 (37, 35-41, 28), SB 139 (131, 123-138, 9), S 72 (65, 56-72, 21), PSB 31 (28, 21- 36, 12), ASB - (116, 11 Li 19, 2), SD - (138, 132-145, 2), PW (excluding ocular sclerites) - (186, 172-193, 7). Scutellum: HS 85 (82, 75-90, 26), LSS 175 (166, 157- 175, 21), cl - (65, 59-71, 14), SS 33 (35, 30-40, 28). Because of distortion of prodorsal sclerite, PS measurement was not made. Gnathosoma (Fig. 7): Palpal setal formula N-N-NNS2-7NB omega (palpal trochanter absent); palpal tibial claw with two distal prongs and basal knob; adoral setae (orl) nude, subcapitular setae (scl) hypertrophied; palpal supracoxal setae (c) absent; cheliceral blade (ch) with single ventral tooth, surrounded by buccal ring (br). Legs (Figs. 1, 3, 4): Femora undivided, six segments beyond the coxal field; pretarsus legs I and II with paired claws and clawlike empodium; pretarsus leg III with normally developed antiaxial claw and claw-like empodium, but with paraxial claw twice as long as antaxial claw. Measurements of holotype with positions of specialized setae given as a ratio of the segment length. Mean, range, and number of paratypes measured given in parentheses. Leg I 200 (196, 188-204, 25); coxal field (Fig. 8) with two setae, one nude 27 (30, 26-36, 13) and other thickened and bifid 19 (18, 17-19, 14); trochanter IB; femur 6B, bv and d setae nude; genu 4B with one seta much longer than others 63 (70, 62-78, 15), two sigma 25 (24, 20-28, 18) and 23 (21, 20-26, 18) at 0.32 (0.35, 0.26-0.47, 25) and 0.57 (0.57, 0.47-0.72, 25), respectively, microseta k 2 (2, 1-2, 14) at 0.84 (0.84, 0.78-0.90, 14); tibia 6B, two phi 22 (19, 15-24, 20) and 16 (14, 11-17, 20) at 0.32 (0.33, 0.26-0.37, 25) and 0.81 (0.80, 0.77-0.84, 25), respectively, k 3 (2, 1-3, 8) at 0.91 (0.87, 0.84-0.89, 8); tarsus 18B, omega 22 (20, 17-23, 22) at 0.21 (0.25, 0.19- 0.38, 25), famulus 3 (2, 1-3, 22) at 0.43 (0.40, 0.36-0.45, 22), two eupathidia 31 376 THE JOURNAL OF ARACHMOLOGY 1 2 3 Figs. 1-5. — Eutrombidium lockleii, new species: 1, holotype leg I; 2, dorsal view of holotype; 3, holotype leg II; 4, holotype leg III; 5, lateral view of paratype prodorsal sclerite. Scale bar 50 jum. See text for explanation of symbols. (30, 25-34, 24) and 13 (14, 1145, 12) at 0.68 (0.70, 0.68-0.77, 25) and 0.86 (0.87, 0.83-0.91, 22), respectively. Leg IL 190 (184, 173-192, 24); coxal field (Fig. 8) with one thick, bifid seta 19 (20, 19-22, 14); trochanter IB; femur 5B, bv and d setae nude; genu 2B with one seta very long 68 (68, 62-76, 12), sigma 17 (21, 15-27, 17) WELBOURN & YOUNG — MITES PARASITIC ON SPIDERS 377 Figs. 6-8. — Eutrombidium lockleii, new species: 6, Scanning electron microscope (SEM) micrograph of engorged paratype (300x); 7, SEM micrograph of ventral gnathosoma (1250x); 8, SEM micrograph of coxal fields legs I and II (1250x). See text for explanation of symbols. 378 THE JOURNAL OF ARACHNOLOGY at 036 (036, 0.28-0.48, 25), k 2 (2, 2-3, 13) at 0.80 (0.75, 0.70-0.81, 10); tibia 5B, two phi 15 (16, 13-21, 19) and 12 (12, 10-14, 14) at 035 (0.34, 0.29-0.39, 25) and 0.78 (0.78, 0.70-0.82, 24), respectively; tarsus MB, omega 18 (18, 17-21, 25) at 0.41 (0.42, 039-0.45, 25), famulus 1 (1, 1-2, 6) at 035 (036, 0.32-0.40, 6), without eupathid. Leg III. 173 (172, 165-192, 23); coxal field with one thick bifid seta 18 (18, 16-20, 13); trochanger IB; femur 4B, bv and d setae nude; genu 2B with both setae very long 64-83 (62-79, 56-89, 24), sigma 25 (21, 17-27, 18) at 038 (0.38, 0.29-0.49, 24); tibia 5B, tarsus 13B, scopa and lophotrix undeveloped. Etymology. — The specific epithet is from the collector’s name, T. C. Lockley. Taxonomic discussion. — Despite the lack of systematic work on North American Eutrombidium , E. lockleii can be easily distinguished from other Eutrombidium in having the ocular sclerites fused into a prodorsal sclerite, eyes on prodorsal sclerite, long barbed seta on palpal tarsus, undeveloped lophotrix and scopa on tarsus leg III, and by the short, rounded idiosoma. It is difficult to assess the relationships of this species with other members of the genus when only six of the 20 named species are known from the larval iestar. Only the discovery of the postlarval instars of this species and rearing of additional Eutrombidium species will allow the relationship of this unusual species to be clarified. SUMMARY OF BIOLOGY All 58 specimens of E. lockleii were obtained from one field 8 km SSW of Indianola, Sunflower Co., Mississippi. This 8 ha hayfield was bordered on the east by a 100 ha fallow pasture, on the north by a 40 ha cotton field, on the west by a deciduous tree-lined wet slough, and on the south by a seasonally dry slough. Coastal bermuda grass predominated, with Erigeron strigosus Muhl. ex. Willd. (Compositae) the most abundant flowering plant during the sampling period. This field is the same as Site #2 of Young and Welbourn (1987), where another new species of mite was discovered attached to tarnished plant bugs (Welbourn and Young 1987). During the period of 12 July to 3 September 1984, 10 vacuum samples were collected weekly at this site, each sample representing 25 row-feet. From these collections, 1530 Ceraticelus emertoni were obtained. Thirty-eight individuals of this species possessed attached larvae of Eutrombidium lockleii (Fig. 9). These collections also contained 208 Oxyopes salticus , of which one individual had two attached larvae of E lockleii. Most specimens of E. lockleii were obtained on 19 July 84, when 35 of 365 C emertoni had mites attached (9.6% parasitization rate). The average body length of C emertoni adults was ca. 1.5 mm, and the average body length of unengorged E. lockleii was ca. 0.2 mm, though some engorged specimens that were still attached exceeded 0.7 mm and were as long as the host prosoma. Multiple attachments did occur, as three spiders were obtained with two mites each, one spider with six mites, and one spider with nine mites attached. Adult and penultimate male and female spiders, as well as small and large immatures, were obtained with attached mites. Two-thirds of the hosts, however, were immature spiders. An analysis of the location of attachment of 56 larval E. lockleii on 40 C. emertoni indicated that all mites were attached along the lines of exuvial WELBOURN & YOUNG— MITES PARASITIC ON SPIDERS 379 Fig. 9. — SEM micrograph of three larval Eutromhidium lockleii new species attached to prosoma of Ceraticelus emertoni (550x). separation (molt sutures) (Fig. 10). This area on each side of the prosoma is also known as the pleuron, a soft and flexible region that allows the stiff carapace and sternum to move in relation to each other (sic “pleurae”; Foelix 1982). More than three-fourths of the mites were located in the median and posterior regions of the pleura (Fig. 11). Attachment to the pleura may be due both to relative ease of cheliceral penetration and to enhanced survivability during host molt. SURVEY OF PARASITIC MITES ON SPIDERS Spiders have a variety of parasites, with most internal forms in the insect orders Diptera, Hymenoptera, and Neuroptera (Eason et al. 1967). Other internal parasites include nematodes which, while rare, are present in a wide range of spiders (Poinar 1985). Mites, on the other hand, are found on the external surfaces and not all are parasitic. While relatively common on certain species (e.g., Parker and Roberts 1974), few mites are reported from spiders in general, perhaps due to difficulties in mite identification. The most frequently encountered mites are phoretic forms, which are usually deutonymphs of the mite suborder Astigmata and are not considered here. Parasitic mites on spiders are reported infrequently, with most species protelean parasites of the prostigmatic cohort Parasitengona. Mites of one mesostigmatic genus have been reported as obligate parasites of spiders. Table 1 summarizes 38 records of parasitic mites associated with spiders of at least 18 families. 380 THE JOURNAL OF ARACHNOLOGY Figs. 10, 11. — Diagramatic views of Ceraticelus emertoni : Lateral, stippled area is the line of ecdysial separation, attachment area for most larvae of Eutrombidium lockleii, new species; dorsal circled numbers represent the percentage of mite attachments to each region. WELBOURN & YOUNG— MITES PARASITIC ON SPIDERS 381 The Trombidiidae account for 16 of the 32 protelean spider parasites, with 11 records of the Holarctic genus Trombidium (Fabricius) on European and North American spiders. Welbourn (1983) reported mites of 10 nominant species from 43 hosts and another 28 hosts with larvae of undetermined Trombidium species. Of these 71 host records, only four were spiders, suggesting that they are accidental hosts for these mites. All records of Trombidium from spiders involve ground strata forms which are more likely than arboreal forms to come in contact with the unengorged mite larvae. Mites of two other closely related trombidiid genera have also been associated with spiders. In Allothrombium, adults of A. lerouxi Moss were reported to attack and kill a Trochosa pratensis (Emerton) (= T. terricola Thorell) spider in Canada (Moss 1960). The larvae of Allothrombium are most often reported from aphid hosts, but there are several records of other arachnid hosts including one from a spider. A second genus, Clinotrombium (Southcott), has two of three named species of mites reported as parasites of spiders in Australia (Southcott 1986). Michener (1946) reported Allothrombium metae Boshell and Kerr parasitizing Pirata spiders in Panama. Examination of Michener’s reared specimens indicates that A. metae should be transferred to Clinotrombium , based on the position of the prodorsal trichobotria and PL setae [= Clinotrombium metae (Boshell and Kerr) new combination]. The second most reported group of mites parasitic on spiders is the Erythraeidae, accounting for 14 of the 32 records. Nearly half of these records are larvae of the cosmopolitan genus Leptus (Latreille). This genus contains approximately 90 named species whose larvae parasitize a wide variety of insect and arachnid hosts. Welbourn (1983) listed 78 arthropod hosts of 30 named Leptus species, and an additional 55 hosts of unidentified Leptus. From those 133 host records, only three species, L. hidakai Kawashima, L. atticolus Lawrence and L. gifuensis Kawashima, are known from spiders. Leptus atticolus and L. gifuensis are known only from the type hosts (spiders) in South Africa and Japan, respectively. Leptus hidakai was found on a spider as well as on opilionids in Japan (Kawashima 1958). Additional collecting and study is needed to determine if these Leptus species are restricted to spiders. The unidentified erythraeid, possibly Leptus , on Diaea sp. (Thomisidae) from New Zealand was pictured by Forster and Forster (1973) and represents the first record from New Zealand. While most protelean parasites are associated with ground-dwelling spiders, Leptus has been found on both aerial and ground-dwelling forms. Two species of Charletonia (Oudemans), C. aranea Southcott and C. miyaxakii (Kawashima), are known only from spiders in India and Japan, respectively, and two new records for the U.S.A. are listed in Table 1. All other species of Charletonia are primarily parasites of Orthoptera and other insects. Lasioery- thraeus Welbourn and Young is a widespread genus in the New World which primarily parasitizes hemipterans, with one record from an immature spider in Mississippi (Young and Welbourn 1987). The new records from Chile (Table 1) represent the southernmost records for the genus. The mesostigmatic family Laelapidae is a large and diverse group which includes free-living predators, arthropod and vertebrate parasites, and nest associates. Mites of the genus Ljunghia (Oudemans) are obligate parasites (non- protelean) of mygalomorph spiders in Indonesia and Australia (Domrow 1975). While all instars can be found on the host, their habits are unknown. This genus 382 THE JOURNAL OF ARACHNOLOGY Table 1. — Parasitic mites on spiders. Parasite Host Country Reference PROSTIGMATA Erythraeidae Charletonia aranea Araneae India Southcott 1966 Southcott C. miyazakii (Kawashima) Theridion sp. (Theridiidae) Japan Kawashima 1958 C. sp. Araneae USA(IL) NEW Philoponeila oweni (Chamberlin) (Uloboridae) USA(AZ) NEW Lasioerythraeus johmtoni Linyphiidae USA(MS) Young & Welbourn 1987 Welbourn & Young L. sp. Cybaeinae (imm.) (Agelenidae) Chile NEW Anyphaenidae (imm.) Chile NEW Leptus atticolus Lawrence Sadis sp. (Salticidae) South Africa Lawrence 1940 L. gifuensis Kawashima Lycosa sp. (Lycosidae) Japan Kawashima 1958 L. hidakai Kawashima Chiracanthium sp. (Clubionidae) Japan Kawashima 1958 L. ignotis (Oudemans) Pachygnatha ciercki Sundeval (Araneidae) England Parker 1962 L. sp. Pardosa sp. (Lycosidae) USA(CT) Sorkin 1982 Philodromus imbecillus Keyserling (Philodromidae) USA(TX) Cokendolpher et al 1979 Undetermined genus Diaea sp. (Thomisidae) New Zealand Forster & Forster 1973 Trombidiidae Allothrombium fuliginosum Lycosa amentata (Hermann) (Clerck) (Lycosidae) England Parker 1965 Clinotrombium antares Linyphiidae Australia Southcott 1986 Southcott C bellator Southcott Salticidae (imm.) Australia Southcott 1986 C. metae (Boshell & Kerr) Pirata sp. (Lycosidae) Panama Michener 1946 (New Comb.) Trombidium poriceps (Oudemans) Araneus diadematus Clerck (Araneidae) Switzerland Andre 1931 Dolomedes fimbriatus Clerck (Pisauridae) Netherlands Oudemans 1912 Linyphia sp. (Linyphiidae) Netherlands Oudemans 1897 Nuctenea umbratica (Clerck) (Araneidae) Switzerland Andre 1931 Zygieila x-notata (Clerck) (Araneidae) Switzerland Andre 1931 T sp. Araneae Canada Welbourn 1983 Agelenopsis sp. (imm.) (Agelenidae) USA(ME) NEW Tegenaria domesticus (Clerck) (Agelenidae) USA(ME) NEW Clubiona moestra Banks (Clubionidae) Canada Welborun 1983 Pardosa hortensis (Thorell) (Lycosidae) Spain Parker & Roberts 1974 WELBOURN & YOUNG— MITES PARASITIC ON SPIDERS 383 Phrurolithus minimus (Koch) (Clubionidae) Spain Parker & Roberts 1974 undetermined genus Neostothis gigas Vellard (Barychelidae) Brasil Vellard 1934 Eutrombidiidae Eutrombidium lockleii, n.sp. Ceraticelus emertoni (Cambridge) (Linyphiidae) USA(MS) NEW Oxyopes salticus Hentz (Oxyopidae) USA(MS) NEW MESOSTIGMATA Laelapidae Ljunghia bristowi (Finnegan) (New Comb.) Liphistius malayanus Abraham (Liphistiidae) Malaysia Finnegan 1933 L. hoggi Domrow Aganippe subtristis Pick.-Camb. (Idiopidae) Australia Domrow 1975 L. pulleini Womersley Selenocsomia stirlingi Hogg (Theraphosidae) Australia Womersley 1956 Aname sp. (Nemesiidae) Australia Domrow 1975 L. rainbowi Domrow Araneae Australia Domrow 1975 L. selenocsomiae Selenocsomia Indonesia Oudemans 1932 Oudemans javanensis (Walck.) (Theraphosidae) (Sumatra) was reviewed by Domrow (1975), where he also redescribed L. selenocosmiae Oudemans from Indonesia. Another mesostigmatic mite from spiders originally named Copriphis ( Pelethiphis ) bristowi Finnegan from Malaysia was placed initially in the Eviphididae. Comparison of Finnegan’s 1933 description with those of Oudemans (1932) and Domrow (1975) indicates that C. bristowi is close to L. selenocosmiae and should be transferred to Ljunghia [= Ljunghia bristowi (Finnegan) new combination ]. ACKNOWLEDGMENTS We appreciate the field and laboratory assistance of T. C. Lockley. Figures 1-5 were inked by P. Brown, Figs. 10 and 11 were prepared by T. C. Lockley. Scanning electron microscope micrographs were prepared by William E. Styre, Ohio State University Agricultural Research Development Center, Wooster, Ohio. Identification of the type host for the newly described species was provided by S. E. Riechert, with some spiders listed in Table 1 identified by N. Platnick, L. Sorkin, and D. T. Jennings. We thank B. M. O’Connor and D. R. Smith (Univ. of Michigan Museum of Zoology) for the Charletonia from an Arizona spider, and N. Platnick (Amer. Mus. Nat. Hist.) for the loan of Michener’s Panama specimens. Travel funds for W. C. Welbourn to study Banks’ material in the Museum of Comparative Zoology (Harvard University) were provided by NSF Grant BSR-8401206. The manuscript also benefited from the reviews of G. L. Bernon, W. A. Bruce, J. C. Cokendolpher, D. E. Johnston, E. E. Lindquist, J. C. Moser, and L. Sorkin. 384 THE JOURNAL OF ARACHNOLOGY LITERATURE CITED Andre, M. 1931. Nouvelles observations sur la larve du Thrombidium holosericeum Linne. Bull. soc. Entomol. France, 1931:259-261. Banks, N. 1896. New North American spiders and mites. Trans. American Entomol. Soc., 23:57-77. Cokendolpher, J. C., N. Y. Homer and D. T. Jennings. 1979. Crab spiders of North-Central Texas (Araneae: Philodromidae & Thomisidae). J. Kansas Entomol. Soc., 54:723-734. Domrow, R. 1975. Ljunghia Oudemans (Acari: Dermanyssidae) a genus parasitic on mygalomorph spiders. Rec. So. Australia Mus., 17:31-39. Eason, R. R., W. B. Peck and W. H. Whitcomb. 1967. Notes on spider parasites, including a reference list. J. Kansas Entomol. Soc., 40:422-434. Finnegan, S. 1933. A new species of mite parasitic on the spider Liphistius malayanus Abraham, from Malaya. Proc. Zool. Soc. London, 1933:413-417. Foelix, R. F. 1982. Biology of Spiders, Harvard Univ. Press, Cambridge, Massachusetts. 306 pp. Forster, R. R, and L. M. Forster. 1973. New Zealand Spiders, an Introduction. Collins, Auckland. 254 pp. Huggans, C. B. and C. C. Blickenstaff. 1966. Parasites and predators of grasshoppers in Missouri. Missouri Agric. Exp. Stn. Res. Bull, 903:1-40. Kawashima, K. 1958. Studies on larval erythraeid mites parasitic on arthropods from Japan (Acarina: Erythraeidae). Kyushu J. Med. ScL, 9:190-211. Lawrence, R. F. 1940. New larval forms of South African mites from arthropod hosts. Ann. Natal Mus., 9:401-408. Michener, C. D. 1946. The taxonomy and bionomics of some Panamanian trombidiid mites. Ann. Entomol. Soc. America, 39:349-380. Moss, W. W. 1960. Description and mating behavior of Allothrombium lerouxi , new species (Acarina: Trombidiidae), a predator of small arthropods in Quebec apple orchards. Canadian Entomol., 92:898-905. Oudemans, A. C. 1897. List of Dutch Acari Latr. Fifth part: Trombidides Leach, with synonymical notes and other remarks, and description of an apparently new, but indeed very old species of Cheyletus, C. squamosus de Geer. Tijdschr. Entomol., 40:117-135. Oudemans, A. C. 1912. Die bis jetzt bekannten Larven von Trombidiidae und Erythraeidae mit besonderer Berkiicksichtigung der fur dee Menschen schadlichen Arten. Zool. Jb., Abt. 1, Suppl. XIV, No. 1, 230 pp. Oudemans, A. C. 1932. Opus 550. Tijdschr. Entomol., 75:202-210. Parker, J. R. 1962. Ectoparasitic mites on spiders. Entomol. Mon. Mag., 98:264. Parker, J. R. 1965. More records of mites as ectoparasites on spiders. British Spider Study Group Bull., 25:6. Parker, J. R. and M. J. Roberts. 1974. Internal and external parasites of the spider Pardosa hortensis (Thorell) (Araneae: Lycosidae). Bull. British Arachnol. Soc., 3:82-84. Poinar, G. O., Jr. 1985. Mermithid (Nematoda) parasites of spiders and harvestmen. J. Arachnol, 13:121-128. Rees, N. E. 1973. Arthropod and nematode parasites, parasitoids and predators of Acrididae in American north of Mexico. U.S.D.A. Tech. Bull, 1460:1-288. Robaux, P. 1974. Recherches sur la developpement et la biologic des Acariens Thrombidiidae. Mem. Mus. natl hist. Nat. Paris, Ser. A, 85:1-186. Severin, H. C. 1944. The grasshopper mite Eutrombidium trigonum (Hermann), an important enemy of grasshoppers. South Dakota Agric. Exp. Stn. Tech. Bull, 3:1-36. Sorkin, L. N. 1982. Parasites of Pardosa wolf spiders (Acarina, Erythraeidae; Insecta, Hymenoptera; Araneae, Lycosidae). American Arachnol, 26:6. Southcott, R. V. 1966. Revision of the genus Charletonia Oudemans (Acarina: Erythraeidae). Australian J. Zool., Suppl, 13:1-84. Southcott, R. V. 1986. Studies of the taxonomy and biology of the subfamily Trombidiidae (Acarina: Trombidiidae) with a critical revision of the genera. Australian J. Zool, Suppl, 128:1-116. Thor, S. and C. Willmann. 1947. Acarina 3 Trombidiidae. Das Tierreich 3, 7 IB: 187-541. Vellard, J. 1934. Notes sur quelques parasites de Mygales Sud-Americaines. Bull soc. Zool France, 59:293-295. Welbourn, W. C. 1983. Potential use of trombidioid and erythraeoid mites as biological control agents of insect pests. Pp. 103-140, In Biological Control of Pests by Mites. (M. A. Hoy, G. L. WELBOURN & YOUNG— MITES PARASITIC ON SPIDERS 385 Cunningham and L. Knutson, eds.). Univ. California (Berkeley) Agric. Exp. Stn. Spec. Publ., 3304:1-185. Welbourn, W. C. and O. P. Young. 1987. A new genus and species of Erythraeinae (Acarina: Erythraeidae) from Mississippi with a key to the genera of North American Erythraeidae. Ann. Entomol. Soc. America, 80:230-242. Womersley, H. 1956. On some new Acarina-Mesostigmata from Australia, New Zealand and New Guinea. J. Linn. Soc., 288:505-599. Young, O. P. and W. C. Welbourn. 1987. The biology of Lasioerythraeus johnstoni (Acarina: Erythraeidae), ectoparasitic and predaceous on the tarnished plant bug, Lygus lineolaris (Hemiptera: Miridae), and other arthropods. Ann. Entomol. Soc. America, 80:243-250. Manuscript received October 1987, revised June 1988. 1988. The Journal of Arachnology 16:387 RESEARCH NOTES AN IRREGULAR ORB-LIKE WEB BUILT BY AN ADULT MALE OF METEPEIRA SP. A (ARANEAE, ARANEIDAE) According to Bristowe (1941), Millot (1949) and Foelix (1982) most males of araneid spiders do not build orb webs after their last molt. However, adult males of Eriophora fuliginea build orb webs (Robinson et al. 1971; Robinson and Robinson 1981). Laboratory studies corroborated that most males of Metepeira sp. A (name suggested by H. W. Levi, in lit.) do not build orb webs (Viera and Costa 1985). The objective of this paper is to report an unusual, irregular web built by an adult male of Metepeira sp. A. In the laboratory, 34 adult males were put into individual glass cages (30 X 30 X 9 cm) with a frame and a water container for 48 h. The temperature averaged 23 ± 2°C, and the photoperiod was 12 h light/ 12 h dark. A specimen of Metepeira sp. A was deposited in the collection of the Museo Nacional de Historia Natural, Montevideo (number 305a). Only one male built one web within this structure: the web was planar, with a vertical diameter of 22.5 cm, a horizontal diameter of 13 cm, several incomplete Fig. 1. — Orb-like web of an adult male Mete- peira sp. A directly drawn from the web. The male built the web in the frame of an experimental cage. Arrows indicate sticky lines. A prey was placed on the sticky lines. 1988. The Journal of Arachnology 16:388 possible radii, and 18 more or less circular sticky lines. Many lines were lax (Fig. 1). One ant (Acromyrmex sp.) was placed onto the sticky lines. The prey stuck but the male failed in its capture. However, this male captured another ant in a female web (Viera and Costa 1985) and also mated normally. This irregular orb like web resembles webs constructed by young Zygiella x- notata (Witt 1956, in Foelix 1982:141) and drugged adult females of Araneus diadematus (Witt 1971). Both drugs and sexual maturity in males modify the expression of the innate program of orb web building. I thank R. Capocasale and F. G. Costa for helpful comments. LITERATURE CITED Bristowe, W. S. 1941,, The Comity of Spiders. II. London: Ray Society. Foelix, R. F. 1982. Biology of Spiders. Harvard Univ. Press. 306 pp. Millot, J. 1949. Ordre des Araneides. Pp. 589-738, In T rake de Zoologie, vol. 6 (P.-P. Grasse ed.), Masson, Paris. Robinson, M. H. and B. Robinson. 1981. Ecologia y comportamiento de algunas aranas fabricadoras de redes en Panama: Argiope argentata , A. savignyi, Nephila davipes y Eriophora fuliginea (Araneae, Araneidae). Acad. Panameea Med. y Cir., 6(1):90-1 17. Robinson, M. H., B. Robinson and W. Graney. 1971. The predatory behavior of the nocturnal orb web spider Eriophora fuliginea (C. L. Koch) (Araneae: Araneidae). Rev. Per. Entom., 14:304-315. Viera, C. and F. G. Costa. 1985. Captura de presas por machos adultos de Metepeira sp. A (Araneae, Araneidae). Actas Jorn. Zool. Uruguay, Montevideo, pp. 5-7. Witt, P. N. 1971. Drags alter web-building of spiders. A review and evaluation. -Behav. Sci. 16(1):98- 113. Carmen Viera, Division Zoologia Experimental, Institute de Investigaciones Biologicas Clemente Estable, Av. Italia 3318, Montevideo, Uruguay. Manuscript received July 1987, revised October 1987 . NORTHERN RECORDS OF MICROBISIUM BRUNNEUM (PSEUDOSCORPIONIDA, NEORISIIDAE) FROM EASTERN CANADA The range of pseudoscorpion species in Canada is poorly known (e.g., Hoff 1958; Dondale 1979; Sharkey 1987). When collecting invertebrates with pitfall traps and by sieving Sphagnum moss in bogs in eastern parts of Canada, 1978 and 1985, the senior author captured the pseudoscorpion Microbisium brunneum (Hagen) both in the boreal forest zone and in northern forestline, forest tundra, areas. M. brunneum was found in samples of Sphagnum moss at the following sites in eastern Canada: 1. Ontario; Copetown (43°14'N, 80°04'W), Summit Hill muskeg, 11 July- 26 September 1978, 2 exx. 1988. The Journal of Arachnology 16:389 2. Quebec; Parc Jacques Cartier, bog at Lac Barette (47°27'N, 71°15'W), 18 July- 14 September 1985, 3 exx. 3. Quebec; Schefferville (54°50'N, 66°50'W), swamp, 21 July 1978, 1 ex. 4. Quebec; Schefferville, open Sphagnum bog, 22 July 1978, 1 ex. 5. Quebec; Kuujjuarapik (Poste-de-la-Baleine) (55° 15'N, 77°50'W), swamp, 9 July - 29 August 1985, 1 ex. 6. Quebec; Kuujjuarapik, paisa bog, 5-28 August 1985, 1 ex. It is worth mentioning that M. hrunneum is the only pseudoscorpion species found at the bogs studied and mentioned above. The habitat fits with the previous data about the ecology of the species: occurring on bogs (Hoff 1946; Sharkey 1987). According to Hoff (1946), M. brunneum has a wide geographical range in eastern Canada and the northern United States. However, no records from eastern Canada (Ontario, Quebec or the Maritime provinces) were included in the list of North American pseudoscorpions by Hoff (1958). Three records have been published for M. brunneum in eastern Canada. Nelson (1984) mentioned the presence of the species in Ontario and Quebec, and Sharkey (1987) in Cape Breton Highlands National Park, Nova Scotia. Kaisila (1964) wrote in his paper on pseudoscorpions collected from Newfoundland in 1949: “ Microbisium sp. (spp.?). These, 1 1 samples in all, constituted the bulk of the material. Finds were made in all parts of the island”. This material, sent to J. C. Chamberlin (Kaisila 1964), probably included M. brunneum. In addition, Hoff (1958) listed M. brunneum just at the forestline area in Churchill, northern Manitoba (about 59° N), based on the report by McClure (1943) as “near M. brunneum” . Besides those specimens collected by the senior author the most northern M. brunneum in the Canadian National Collection was taken 40 miles west of St. John’s, Newfoundland ex muskeg at about 47°50'N. The present samples of M. brunneum from Schefferville and Kuujjuarapik are clearly the northernmost known in the eastern part of Canada. Although the latitude of these sites is more southern than that of Churchill, the environmental conditions are comparable: all these three areas are situated in the forestline region or forest tundra (see e.g., Danks 1981). M. brunneum is not the pseudoscorpion with the most northern distribution in North America. An undescribed species of Wyochernes (presently being described by W. B. Muchmore) was discovered by V. Behan-Pelletier in the Yukon Territory at the following locality: British Mountains, 350 m, Sheep Creek, 69° 10'N, 140° 18'W, 23 June 1984, collected under stones on fine gravel about 1 m from edge of creek. The generous help from the Centre d ’etudes nordiques, Universite Laval (Quebec City), and from the Schefferville Subarctic Research Station, McGill University (Montreal), during the field work of the senior author is greatly acknowledged. LITERATURE CITED Danks, H. V. 1981. Arctic arthropods. A review of systematics and ecology with particular reference to the North American fauna. Entomol. Soc. Canada, Ottawa. 608 pp. Dondale, C. D. 1979. Opiliones, Pseudoscorpionida, Scorpionida, Solifugae. Pp. 250-251, In Canada and its insect life (H. V. Danks, ed.). Mem. Entomol. Soc. Canada, No. 108. 1988. The Journal of Arachnology 16:390 Hoff, C. C. 1946. American species of the pseudoscorpion genus Microhisium Chamberlin, 1930. Bull. Chicago Acad. Sci., 7:493-497. Hoff, C. C. 1958. List of the pseudoscorpions of North America north of Mexico. American Mus. Novitates, No. 1875:1-50. Kaisila, J. 1964. Some pseudoscorpionids from Newfoundland. Ann. Zool. Fennici, 1:52-54. McClure, H. E. 1943. Aspection in the biotic communities of the Churchill area, Manitoba. Ecol. Monogr., 13:1-35. Nelson, S. Jr. 1984. The genus Microhisium in North and Central America (Pseudoscorpionida, Neobisiidae). J. Arachnol., 12:341-350. Sharkey, M. J. 1987. Subclass Chelonethida, Order Pseudoscorpionida (pseudoscorpions). P. 17, In The Insects, Spiders and Mites of Cape Breton Highlands National Park. Agric. Canada, Biosystematics Res. Centr., Rep. 1. Seppo Koponen, Zoological Museum, University of Turku, SF-20500 Turku, Finland, and Michael J. Sharkey, Biosystematics Research Centre, Agriculture Canada, Ottawa, Ontario K1A 0C6, Canada. Manuscript received March 1988, revised April 1988. PREDATION OF ACHAEARANEA TEPIDARIOR UM (ARANEAE, THERIDIIDAE) UPON SPHODROS FITCHI (ARANEAE, ATYPIDAE) Sphodros fitchi Gertsch and Platnick is a recently described purseweb spider inhabiting the central plains states from Nebraska to Oklahoma and Arkansas (Gertsch and Platnick 1980). Although some aspects of the natural history of members of this genus have been observed (Coyle and Shear 1981; McCook 1888; Morrow 1985; Teeter 1984), little information exists concerning predation. A female Sphodros rufipes (Latreille) was taken from the stomach of a frog (Gertsch 1936). Observations in eastern Kansas indicate that males of the same species often fall victim to female conspecifics and females of Sphodros niger (Hentz) during the mating season (Morrow 1985). The present note records predation of Achaearanea tepidariorum (C. L. Koch) upon S. fitchi . On 10 July 1987, remains of an adult male S. fitchi were discovered in the web of a female house spider, A. tepidariorum , located in a metal storage building on the University of Kansas Rockefeller Experimental Tract in Jefferson County, Kansas. The web was situated below a wooden shelf against a wall, and was approximately 0.5 m above the concrete floor. The Sphodros was wrapped in silk and suspended in the lower portion of the web. A. tepidariorum is well known for its ability to overpower and consume relatively large prey, including vertebrates (Gertsch 1979). Due to the shriveled condition of the abdomen, the total length of the victimized Sphodros was not measured; however, the length of the carapace was 4.1 mm. Since the male holotype of this species has a carapace length of 4.2 mm and a total length of 12.7 mm (Gertsch and Platnick 1980), the estimated length of the prey item is less than 13 mm. The total length of the female Achaearanea was 7.4 mm. 1988. The Journal of Arachnology 16:391 Upon reaching maturity, Sphodros males emerge from their burrows and wander in search of suitable mates (Coyle and Shear 1981). During this period, they are especially vulnerable to predation. Fitch (1963) observed a jumping spider, Phidippus audax (Hentz) (Salticidae), attack and quickly kill a male S. fitchi that was confined in an open glass jar in his laboratory. In view of an interesting account of a trapdoor spider ( Ummidia sp.) (Ctenizidae) caught by a Steatoda triangulosa (Walckenaer) (Horner and Russell 1986), S. triangulosa and other theridiids could conceivably prey upon male Sphodros. I thank Dr. Norman Platnick of the American Museum of Natural History for spider identifications, Dr. Charles Michener, University of Kansas, for providing laboratory space, and Paul Liechti, Kansas Biological Survey, for providing a microscope and supplies. For reviewing the manuscript I thank: Dr. George Byers, Dr. Henry S. Fitch, and Joseph T. Collins, University of Kansas, and Dr. Norman Platnick. Specimens were deposited in the American Museum of Natural History. LITERATURE CITED Coyle, F. A. and W. A. Shear. 1981. Observations on the natural history of Sphodros ahboti and Sphodros rufipes (Araneae, Atypidae), with evidence for a contact sex pheromone. J. Arachnol., 9:317-326. Fitch, H. S. 1963. Spiders of the University of Kansas Natural History Reservation and Rockefeller Experimental Tract. Univ. Kansas Mus. Nat. Hist. Misc. Pub., 33:1-202. Gertsch, W. J. 1936. The nearctic Atypidae. Amer. Mus. Nov., No. 895, 19 pp. Gertsch, W. J. 1979. American Spiders (2nd edition). Van Nostrand Reinhold Co., New York, 274 pp. Gertsch, W. J. and N. I. Platnick. 1980. A revision of the American spiders of the family Atypidae (Araneae, Mygalomorphae). American Mus. Nov., No. 2704, 39 pp. Horner, N. V. and D. Russell. 1986. Ummidia trapdoor spider caught in a Steatoda web (Araneae: Ctenizidae, Theridiidae). J. Arachnol., 14:142. McCook, H. C. 1888. Nesting habits of the American purseweb spider. Proc. Acad. Nat. Sci., Philadelphia, 203-220. Morrow, W. 1985. Two species of atypid spiders (Araneae, Atypidae) in eastern Kansas: male emergence times and notes on natural history. Masters Thesis Univ. Kansas, 47 pp. Teeter, M. M. 1984. The role of slope orientation in nest-site selection by Sphodros spp. (Araneae, Atypidae): field and experimental observations. Masters Thesis Univ. Kansas, 40 pp. Hank Guarisco, P. O. Box 3171, Lawrence, Kansas 66046 USA. Manuscript received March 1988, revised May 1988. COMMENTS ON A WOLF SPIDER FEEDING ON A GREEN ANOLE LIZARD Reports of terrestrial, araneomorph spiders feeding on vertebrates are infrequent. Cokendolpher (1977. J. Arachnol., 5:184) observed a female Argiope aurantia Lucas eating a Eumeces laticeps Schneider (broad-headed skink). The present note is the first report of a wolf spider feeding on a green anole. 1988. The Journal of Arachnology 16:392 On 19 February 1988 at 0700 hours, I observed a male Lycosa ammophila Wallace feeding on an Anolis carolinensis carolinensis Voight (family Iguanidae). The predation occurred in a sandhill community at Wekiwa Springs State Park, Wekiwa Springs, Orange County, Florida. Dominant trees in the sandhill community are longleaf pine, Pinus palustris Mill., and Turkey oak, Quercus laevis Walt. The understory is dominated by wiregrass, Aristida stricata Michx., and saw palmetto, Serenoa repens (Bartr.) Small. A. carolinensis carolinensis is an abundant lizard in this region of the south, found on trees, shrubs, vines, or the ground. The attack on the anole was not observed. The spider may have encountered the anole at night while it was asleep. I found the spider and anole after opening the back of a Sherman small mammal trap. It is not known if the spider cornered the lizard in the trap or dragged the lizard into the trap after catching it. The anole measured 3.9 cm (snout-vent length) and was found in the chelicerae of the spider. The anole received two bites which penetrated the body. The first was immediately behind the hind limbs and the second bite was immediately behind the right fore limb. The wolf spider’s jaws were located in the second bite area when first observed. The spider dropped the anole, dropped off the door of the trap and disappeared in the litter. A search of the litter failed to produce the spider. Identification of the spider is based on (1) size, (2) coloration, and (3) the fact the site has been sampled for 18 months with L. ammophila being the only large wolf spider of that size and coloration collected. Lycosa ammophila is a large spider belonging to the lent a group (Wallace, 1942. American Mus. Nov. No. 1 185:1-21). It appears that L. ammophila is large enough to handle a medium sized anole without any harm coming to itself. This work was supported by Nongame Wildlife Program contract no. RFP-86- 003 from the Florida Game and Ffesh Water Fish Commission. David T. Corey, Department of Biological Sciences, University of Central Florida, Orlando, Florida 32816 USA. Manuscript received March 1988, revised May 1988. NOTES SUR LE DEVELOPPEMENT POSTEMBRYONNAIRE DE TITYUS STRANDI (SCORPIONES, BUTHIDAE) Parmi les Scorpions collectes dans la region de Tucurui, Etat de Para, Bresil (Lourengo, W. R., sous-presse, Bol. Mus. par. E. Goeldi), quelques exemplaires ont ete preleves vivants parmi lesquels une femelle de Tityus strandi Werner, 1939, qui s’est reproduite au Laboratoire a Paris, donnant naissance a 3 portees successives, sans nouvelle fecundation, phenomene deja observe chez des especes du genre Tityus telles T. bahiensis (Matthiesen, F. A., 1970, Bull. Mus. natn. Hist, nat., Paris, 2e ser., 41(6): 1367-1370) et T. fasciolatus (Lourengo, W. R., 1979, Rev. nordest. Biol., 2(l/2):49-96). 1988. The Journal of Arachnology 16:393 Fig. 1 — Distribution des valeurs mor- phometriques (en mm), pour les stades juveniles et adulte chez Tityus strandi. LPr = longueur du prosoma; LAs - longueur du cinquieme anneau du metasoma. Chaque point represente au moins un individu. Les connaissances sur la biologie du developpement des Tityus d’Amazonie sont encore tres incompletes, et nous rapportons ici les quelques observations faites sur le developpement de Tityus strandi (Fig. 1). La femelle etudiee a produit des portees les 28 janvier 1985 (13 petits), 5 juillet 1985 (12 petits), et 25 octobre 1985 (11 petits). Les durees du developpement embryonnaire sont done de 158 jours et 86 jours, valeurs voisines de celles observees pour Tityus fasciolatus. Les petits des 3 portees passent la premiere mue 4 jours apres leur naissance. Seuls 4 individus de la premiere portee et 5 de la deuxieme portee, muent une deuxieme fois. Les quatre premiers dans la periode du 13 au 15 juin 1985 (a 137 jours en moyenne) et les 5 individus de la deuxieme portee entre le 2 et le 5 novembre 1985 (soit a 120 jours en moyenne). Pour la troisieme portee, les observations ont ete plus completes. Tous les 11 individus passent la 2eme mue entre le ler et le 8 janvier 1986 (a 71 jours en moyenne). Six individus passent la 3eme mue entre le 19 mars et le 15 avril 1986 (entre 145 et 172 jours). La quatrieme mue est observee pour quatre individus entre le 13 et le 18 mai 1986 (entre 200 et 205 jours). Finalement le stade adulte est acquis avec la 5eme mue pour quatre individus, 3 femelles et 1 male entre le 30 septembre et le 17 novembre 1986. Ainsi, la duree du developpement postembryonnaire se situe entre 340 et 388 jours. La progression de la croissance est exprimee dans le graphique 1, a partir des valeurs relevees sur les individus morts et les exuvies, males et femelles confondus. Sont pris en consideration la longueur de la plaque prosomienne (LPr) et la longueur du 5eme anneau metasomal (LA5). Les resultats obtenus sur le developpement de Tityus strandi sont voisins de ceux obtenus pour d’autres Buthidae neotropicaux. Wilson R. Lourengo, Laboratoire de Zoologie (Arthropodes), Museum National d’Histoire Naturelle, 61, rue de Buffon 75005 Paris, France; and Vera Regina D. von Eickstedt, Institute Butantan, Segao de Artropodos Pegonhentos, 05504, Sao Paulo, Brasil. Manuscript received March 1988, revised June 1988. 395 THE JOURNAL OF ARACHNOLOGY Instructions for Authors GENERAL COMMENTS These instructions should be consulted carefully to avoid unnecessary delays in the review, editing, and printing phases of the publication process. If in doubt, authors should examine the style used in recent issues of The Journal of Arachnology and/or consult the Associate Editor. Typescripts are preferred in English but are also acceptable in Spanish, French, or Portuguese. They must be TYPED DOUBLE- OR TRIPLE-SPACED THROUGHOUT. Good quality paper should be used and erasable bond is specifically excluded. Leave margins at least 1.5 in. (4 cm) on the left and 1 in. (2.5 cm) on the top, bottom, and right. Do not hyphenate any words at the right margin. Do not edit your own typescript. Italics are permitted only to indicate scientific names; only the TITLE, PRIMARY HEADINGS (e.g., INTRODUCTION, etc.) and RUNNING HEAD should be typed in capital letters. Follow the recommendations of Council of Biological Editors (CBE) Style Manual , Third Edition, unless indicated otherwise below. To facilitate prompt review by two or more referees, send to the Associate Editor three identical copies of the typed material, one of which should be designated as the “original”, together with photocopies of the illustrations. Original artwork or PMTs may be retained by the author until the completion of the review process and then sent with the revised typescript, or if no revision is necessary, sent at the request of the Associate Editor. Typescripts longer than about 1500 words (five double-spaced typewritten pages in Elite, six double- spaced typewritten pages in Pica), or the equivalent in printed words plus artwork, should be prepared as Feature Articles, shorter papers as Research Notes. Typescripts and all related correspondence must be sent to the Associate Editor: Dr. Jerome S. Rovner, Department of Zoological Sciences, Ohio University, Irvine Hall, Athens, Ohio 45701 USA. FEATURE ARTICLES Arrange the various parts of feature articles in the following sequence: (1) mailing address, (2) title, (3) by-line, (4) abstract, (5) body of text, (6) acknowledgments, (7) literature cited, (8) figure legends, (9) footnotes, (10) running head, (11) tables with legends, (12) figures. Put only parts 1, 2 and 3 on page 1 , and start page 2 with part 4. All other pages are numbered consecutively. Mailing address. — Include the complete address and telephone number of that author with whom all correspondence with the Editors should be handled. Title. — When using common or scientific names in the title, be sure to include in parentheses the appropriate higher taxa (order, family,...), but exclude the author of the scientific name. Include footnote indication if appropriate (e.g., to acknowledge grant support), but type footnote separately (see part 9 above). The title should be typed in capital letters and may not exceed 48 characters and spaces per line. By-line. — Include the name(s) of author(s) as customarily written (less titles) and complete address(es), including zip code or other postal designation. Include footnote indication(s) if appropriate (e.g., to indicate change of address), but type footnote separately (see part 9 above). Leave three spaces between title and by-line, and four spaces between name(s) and address(es). Abstracts. — The abstract should be a summary of the basic findings, and should not exceed 2 to 3 percent of the text in length. Papers in a language other than English must be accompanied by an English abstract as well as an abstract in the language of the text. Papers written in English but focusing on a subject pertaining to a country where another of the accepted languages is used, may include an abstract in that language. Body of text. — Use whatever form seems best to accommodate the necessary description of research. Be concise. Numerals: Use the metric system for all measurements (the English system is acceptable only when transcribing locality data accompanying museum specimens). Write out numbers from one to nine in words unless: (1) there are decimals, or (2) if there is a series of numbers, some of which are 10 or greater, or (3) if they precede standard units of measurement. Time of day should be indicated using the 4-digit, 24-hour clock style without punctuation and with “hours” spelled out. Use the period rather than the comma as a decimal point. 396 Abbreviations: Conform to standard references for abbreviations, i.e., the CBE Style Manual for scientific abbreviations, and a good language reference. Note that abbreviations of metric units of measurement are not punctuated (e.g., mm and km, but ft. and mi.). Some common abbreviations should be typed as follows: h (hours), min (minutes), s (seconds). Most statistical and arithmetic symbols are italicized: P (probability), N (sample size), df (degrees of freedom), /-test; but SD (standard deviation), SE (standard error). Years should be written in full, e.g., 1988-1989, not 1988-89. Write out “male” or “female” rather than use symbols. Citations: Citations within the text should be in the following form: Bellrose (1950); Bellrose (1950:33); or (Bellrose 1950), unless the citation is for the author of a scientific name, and then use a comma between name and date (Bellrose, 1950). The complete scientific name of a species or genus of arthropod, including author(s), must be given the first time it is mentioned in the text. Use single-line notation for fractions [e.g., 1/4 and not '/4; (4- 1 2)/ 3 and not -p]. Whenever an author indicates an unnamed taxon, (e.g., A-us sp. or A-us sp. A), there must be a designated voucher specimen deposited in a recognized institution and a statement indicating the location of this specimen. It is also recommended that voucher specimens be designated for taxa whose identity may be difficult or uncertain. Taxonomic papers: The following special directions apply to authors of taxonomic papers: (a) . Do not use abbreviations in a primary heading to indicate that a new name or a new combination is being proposed (e.g., A-us-x-us, new species, rather than A-us x-us, n. sp. or comparable abbreviations). (b) . Keys must be typed as follows: 1. Arabic numerals to designate the leading entry of a couplet 2 Do not designate the second entry of a couplet, either by means of numbers, letters or other marks A-us x-us 2. Type numbers flush to left margin, and start entry on fifth space A-us y-us Subsequent lines of any entry must also be indented five additional spaces as in this example of an entry with two or more printed lines A-us z-us (c) . Synonymies must follow the abbreviated style shown below: A-us x-us Jones, 1930:3, 1935:9; Russell 1945:453; Smith 1954a: 16, 1954b:678; Cooper and Lim 1955:18 (in part). A-us y-us Bates, 1932:18, fig. 4. NEW SYNONYMY. A-us z-us : Miranda 1948:98 (misidentification): Harris 1951:3 (in part?), {nec A-us z-us Zimmer), (d) . Lists of specimens examined of a given taxon must be the last items typed in the treatment of that taxon as they will be set in smaller type. Adhere to the following style for listing specimens examined: Country: State or comparable political subdivision; County or District, detailed locality (elevation), 14 July 1945 (collector), 2 males, 5 females (acronym of institution where specimens are deposited), next detailed locality within that County, etc.; next County in the same State; etc.: next State in the same Country: etc. Next Country: etc. Punctuation rules are very simple. Use a period to separate countries, a colon to separate states, a semi-colon to separate counties, and a comma to separate specific localities. Acknowledgments. — Avoid overlooking persons who have in some substantial way assisted with the work. Authors of taxonomic papers should spell out the name and indicate parenthetically the acronym of institutions where specimens studied are deposited, if not mentioned elsewhere in the text. Literature cited. — Include only those publications to which reference is made in the text. Adhere to the CBE Style Manual and refer to a previous issue of The Journal of Arachnology for style. Do not abbreviate place names in journal citations. Repeat name(s) of author(s) in case of multiple entries. Figure legends. — Provide one legend for each illustration which will be reproduced separately, or for each “plate” consisting of several illustrations. Adhere to the following styles: Figures 1-4. — A-us x-us, male from Timbuktu: 1, left leg; 2, right chelicera; 3, dorsal aspect of genitalia; 4, ventral aspect of abdomen. Figures 27-34. — Right chelicerae of species of A-us from Timbuktu: Figs. 27, 29, 31, 33. — Dorsal views; Figs. 28, 30, 32, 34. — Prolateral views of movable finger; 27, 28, A-us x-us, holotype male; 29, 30, A-us w-us, male; 31, 32, A-us z-us, holotype male; 33, 34, A-us t-us, male. Scale = 1.0 mm. Type all figure legends consecutively on the same page(s), double spacing lines within each legend and leaving 4 spaces between legends. Keep in mind that 99 characters and spaces represent one 397 printed line, and that 4 mm for each printed line must be subtracted from the maximum height permissible on full-page illustrations. Footnotes.— Footnotes will permitted on page 1 when it is appropriate to acknowledge grant support and to indicate change of address, etc. No footnotes will be permitted on any other page. Type footnotes together on a separate page. Running head. — At the top of each right hand page will be printed the author(s)’ last name(s) and an abbreviated title. This running head may not exceed 60 characters and spaces, must be typed in capital letters, and a hyphen must separate the name(s) from the title. Place the running head on a separate page. Tables with legends. — Prepare all tables precisely as they are to be typeset and include the legend at the top of each. Only in special cases will tables be photo-reproduced rather than typeset. Construct tables as simply as possible. Make marginal notations in the text which clearly indicate the appropriate point of insertion of each table. Note that in these instructions “graphs” are regarded as “Figures,” not “Tables.” The size of the printed page imposes a limit on the size of tables that can be accepted. In a normal, upright table these limits are 99 characters and spaces per line and 56 lines in height, including legend. Sideways tables are difficult to typeset and present problems during paste-up; therefore, they will be accepted in exceptional cases only, as most tables that are “too wide” can usually be rearranged to fit into the more desirable, upright position. Any table that exceeds the dimensions given above may not be accepted for publication. Tables must be typed double-spaced throughout, and the legend must be in the following style: Table 1. — Incorporate into the legend all information necessary to make the data presented clearly understandable. Meanings of symbols can be also included in the legend, (e g , a = sampled In 1988, b = sampled in 1989). Do not use footnotes. Figures. — Figures are the only parts of the paper for which the author is responsible for preparation of camera-ready material, and therefore they must be executed with great care. Drawings should be made with black drawing ink on high quality, smooth, white, heavy paper. Drawings made in some manner not employing black ink (e.g., shaded pencil drawings) are more difficult to print than photographs. When such a technique is used, the drawings should be fairly dark, have good contrast, and be well executed. The largest dimensions on the printed page on which illustrative material (with legends) will be printed is 127 by 200 mm. Decisions as to size of originals, placement of several smaller drawings on a single page, etc., should be based upon this format. Where several smaller drawings are to appear together on a single page, it is the responsibility of the author(s) to furnish an assembled “plate” precisely as it should appear in print. An effort should be made to efficiently utilize the available space within the limits of the original “plate”. The “plate” must be prepared to allow for inclusion of the legend beneath the drawings on the same printed page. Number all drawings consecutively using Arabic numerals; do not designate illustrations with letters (except in special cases such as behavioral sequences), and never designate plates. Labeling should be neat, secure, and done with a method which produces flat characters, e.g., pressure-sensitive transfer characters or a lettering apparatus employing black ink. Photographs should have good contrast, be glossy or semiglossy, and printed slightly larger (up to 2X) than anticipated published size. Photographs and inked drawings should not be grouped together in a single plate, because they will be difficult to print. Illustrative material of any kind larger than two times the anticipated printed size may not be accepted. Larger illustrations are difficult to handle and are easily damaged or lost in the mail. Authors preparing larger illustrations should consult local photo shops and print firms about the availability of PMTs (PhotoMechanical Transfer) for reducing their illustrations to an acceptable size. PMTs are high contrast, direct positive-to-positive reproductions on semi-glossy paper, which will also give the author(s) a good idea of how the illustrations will look when printed. PMTs can be obtained at a reasonable cost in a number of sizes, usually 8 by 10 in. or 9 by 12 in., with any amount of figure reduction desired within that space. Sending PMTs rather than originals avoids the loss of art work in the mail, and if the PMTs get lost or damaged they can be replaced. Illustrations, PMTs, and photographs must be securely mounted on stiff white paper or mounting board (Scotch Magic Transparent Tape can be used on the white portions of illustration paper). On the back of each plate include author’s name(s), abbreviated title of typescript (or running head), figure numbers included, and orientation (= topside). Illustrative material that is poorly prepared, too large, or not mounted properly may be returned to the author. Although every effort will be made to return artwork if requested, safe return cannot be guaranteed. 398 Make marginal notations in the text which clearly indicate the appropriate points of insertion of all figures. RESEARCH NOTES Arrange the various parts of your research note in the following sequence: (1) mailing address, (2) title, (3) body of text, (4) by-line, (5) figure legends, (6) tables with legends, and (7) illustrations. Follow instructions given above for feature articles unless otherwise indicated below. Do not include footnotes anywhere. References to grant support and all other acknowledgments should be made in a statement in the body of the text. If an indication of change of address is desired, it should be included parenthetically after the credited institution as “present address.” For citation of literature, follow the same instructions as for feature articles. The by-line must be typed in paragraph form after the body of the text or the literature cited when one is present. When possible, Research Notes will take priority over Feature Articles in the publication sequence. THE AMERICAN ARACHNOLOGICAL SOCIETY President: William A. Shear (1987-1989) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 Membership Secretary: Norman L Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: James W. Berry (1987-1988) Department of Zoology Butler University Indianapolis, Indiana 46208 Directors: James C. Cokendolpher (1987-1989), W. G. Eberhard (1986-1988), Jerome S. Rovner (1987-1989). Honorary Members: P. Bonnet, W. J. Gertsch, H. Homann, R. F. Lawrencef, H. W. Levi, G. H. Locket, A. F. Millidge, M. Vachon, T. Yaginuma. The American Arachnological Society was founded in August, 1972, to promote the study of Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachnology. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $30.00 for regular members, $20.00 for student members and $50.00 for institutions. Correspondence concerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addition, members receive the bi-annual newsletter of the Society, American Arachnology. American Arachnology, edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-find literature, information about arachnology courses and professional meetings, abstracts of papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arachnology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. President-Elect: George W. Uetz (1987-1989) Department of Biological Sciences University of Cincinnati Cincinnati, Ohio 45221 Treasurer: Gail E. Stratton (1987-1989) Department of Biology Albion College Albion, Michigan 49224 Archivist: Vincent D. Roth Box 136 Portal, Arizona 85632 f Deceased Research Notes An irregular orb-like web built by an adult male of Metepeira sp. A (Araneae, Araneidae), Carmen Viera 387 Northern records of Microbisium brunneum (Pseudoscorpionida, Neobisiidae) from eastern Canada, Seppo Koponen and Michael J. Sharkey 388 Predation of Achaearanea tepidariorum (Araneae, Theridiidae) upon Sphodros fitchi (Araneae, Atypidae), Hank Guarisco 390 Comments on a wolf spider feeding on a green anole lizard, David T. Corey 391 Notes sur le developpement postembryonnaire de Tityus strandi (Scorpiones, Buthidae), Wilson R. Lourengo 392 Other Instructions for Authors 395 CONTENTS THE JOURNAL OF ARACHNOLOGY VOLUME 16 Feature Articles NUMBER 3 Revision of the genus Pycnothele (Araneae, Nemesiidae), Fernando Perez- Miles and Roberto M. Capocasale 281 Behavioral flexibility in orb web construction: Effects of supplies in different silk glands and spider size and weight, William G. Eberhard 295 Orb web recycling in Araneus cavaticus (Araneae, Araneidae) with an emphasis on the adhesive spiral component, GABamide, Mark A. Townley and Edward K. Tillinghast 303 Sexual behavior in Dictyna volucripes (Araneae, Dictynidae), Christopher K. Starr 321 Spider fauna of flooded rice fields in northern California, Michael J. Oraze, Albert A. Grigarick, Joseph H. Lynch and Kirk A. Smith 331 Four new species of Paratheuma (Araneae, Desidae) from the Pacific, Joseph A. Beatty and James W. Berry 339 Factors influencing specificity and choice of host in Argyrodes antipodiana (Theridiidae, Araneae), Mary E. A. Whitehouse 349 Xenonemesia , un nuevo genero de Nemesiidae (Araneae, Mygalomorphae), Pablo A. Goloboff 357 Typhlochactas mitchelli , a new species of eyeless, montane forest litter scorpion from northeastern Oaxaca, Mexico (Chactidae, Superstitioninae, Typhlochactini), W. David Sissom 365 Mites parasitic on spiders, with a description of a new species of Eutrombidium (Acari, Eutrombidiidae), W. Calvin Welbourn and Orrey P. Young 373 (continued on back inside cover) Cover photograph, spider figure on men’s meeting house, Palau, by J. W. Berry Printed by the Texas Tech Press, Lubbock, Texas, U.S.A. Posted at Lubbock, Texas, January 4, 1989 ) SM II HSON IAIN IN b I i I U ll u in imuu.iuuoimi_i\iv iNuanni^b b d i d V d ti II LIBRARItb SMITHSONIAf ) ^ ___ 00 z \ ^ ^ . co J ^ LiJ CO W ^ UJ < (ie. 3“ a < 0 5 x>^fr 2 5 ^ZCauksp/ ^ * / 2 „J 2 ~J z NVIN0SH1IWS S3 I B V8 9 1“1 LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNi NVIN0SH1IIM 1 m /£§^\ - , . I m " J 9 ^ m _ co — co «_ (/) SMITHSONIAN INSTITUTION NOliniliSNi NVIN0SH1IINS S3iavaail LIBRARIES SMITHSONIAI o Ofe*- ^21 mA7\ ^ rJL«M -r \*W. x>^^/ o X . J/ ? > * W3” 2 ^ 2 X > CO ^ 2 CO 2 CO * 2 NVIN0SH1IWS S3 I BVd 9 11 LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNi NVIN0SH1IW CO =; CO “ CO — O X DS>2 o n^u)^/ q SMITHSONIAN- 'INSTITUTION NOlirUllSNlAlVINOSHilWS2^ I a V a 8 IT L I B R A R I ESZ SMITHSONIA m ' Vj^sr X, m co “ co _ co rz nvinoshii/js S3 lava an libraries Smithsonian institution NOliniliSNi nvinoshiiw co 2 \ CO 2 V CO 2 ^%os\^>£/ > ’ 5 V*P' > ^ \v > 2 CO ‘ 2 cO 2 CO SMITHSONIAN INSTITUTION NOIlfUliSNI NVINOSHillNS S3ldVd8IT LIBRARIES SMITHSONIAI co -* \ co n co NviNosHiiiAjs S3iavaan libraries Smithsonian institution NOliniliSNi nvinoshim ... 2 ^ ^ i- ^ 2 r- — ^ 2 r- 9 s<&E^ Nv ™ 9 ,ro 9 m *«u t > CO m co r= co — co SMITHSONIAN INSTITUTION NOliniliSNi NVIN0SH1IWS S3iavaaiT LIBRARIES SMITHSONIA T/i 2 CO 2 .v. CO > > *‘W’~ 2 X > CO ~ 2 CO 2 co * 2 NVINOSHilWS S3 lava a IT LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNi NVINOSHilW co ~ co =; co x CO _ o SMITHSONIAN”1 INST I TUT I ON NOliniliSNi “NVINOSHIUNS S3 I ava 8 IT LIBRARIES SMITHSONIA : r- 2 i— 2 r- 2 ) . /fOa ,Vobv O /fdsT IT/TIn. _ w O 2*IsTI0/bv 2^7 0 O ArrfSfr* . TITUTION NOIlflillSNI NVINOSHIHNS S3iavaan LIBRARIES SMITHSONIAN INSTITUTION avaan libraries Smithsonian institution NoumusNi nvinoshiiws saiavaan i ° , . voo 2 w I co m m co — co — tn *•. zz in TITUTION NOlifUllSNI NVINOSHIHNS S3 I avaan LIBRARIES SMITHSONIAN INSTITUTION I v> z ^ ^ 5 £ z co s •- ^ , <; - .4X s 5 /53I2X ~ 4v> s. s 1 s&AF s «k>- * ier 'hk o likVv x > > ‘W -i ~\ x^uiiv^ ^ > z co 2: co *■■ z co *■• z avdail LIBRARIES SMITHSONIAN INSTITUTION NOIlflillSNI NVINOSH1IIAIS S3ldVdail l ^ in ^ ^ ^ 00 ^ ..v (O — CO ^ < co ^ m ^ TITUTION NOIlflillSNI NVIN0SH1IIAIS SaiBVBail^LIBRARI ES~" SMITHSONIAN”1 INSTITUTION Z f' avaan libraries Smithsonian institution NoumusNi nvinoshihns saiuvaan l 2: „ in z v co z co z > ' yV > ' X^VQSV^X > co 2 co ■*' z co 2: co NTUTION NOIlflillSNI NVIN0SH1IWS S3ldVUan LIBRARIES SMITHSONIAN INSTITUTION ^ nvaaii libraries Smithsonian institution NoumusNi nvinoshihns saiavaan i r* z r~ z O ~ " y > / ra r 33 > = % m — "'<000* m m co ± co £ — co _ TITUTION NOIlflillSNI NVIN0SH1IWS S3ldVaail LIBRARIES SMITHSONIAN INSTITUTION co z co z .»«• . co z v- co 2 < . 2 < in W § CO z CO v. Z CO — avaan libraries Smithsonian institution NoumusNi nvinoshiiws saiavaan 1 “ CO - CO — CO — CO TITUTION NOIlflillSNI NVIN0SH1IWS S3ldVaail LIBRARIES SMITHSONIAN INSTITUTION N Z r~ Z r- Z r~ Z