' ’<1 li i Q.L- The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY JUN 1 9 1997 ii/BRARitS VOLUME 25 1997 NUMBER 1 THE JOURNAL OF ARACHNOLOGY EDITOR: James W. Berry, Butler University ASSOCIATE EDITOR: Petra Sierwald, Field Museum EDITORIAL BOARD: A. Cady, Miami (Ohio) Univ. at Middletown; J. E. Carrel, Univ. Missouri; J. A. Coddington, National Mus. Natural Hist.; J. C. Cokendolpher, Lubbock, Texas; F. A. Coyle, Western Carolina Univ.; C. D. Dondale, Agriculture Canada; W. G. Eberhard, Univ. Costa Rica; M. E. Galia- no, Mus. Argentino de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia, Missouri; C. Griswold, Calif. Acad. Sci.; N. V. Horner, Midwestern State Univ.; D. T. Jennings, Garland, Maine; V. F. Lee, California Acad. Sci.; H. W. Levi, Harvard Univ.; E. A. Maury, Mus. Argentino de Ciencias Naturales; N. I. Plat- nick, American Mus. Natural Hist.; G. A. Polis, Vanderbilt Univ.; S. E. Riechert, Univ. Tennessee; A. L. Rypstra, Miami Univ., Ohio; M. H. Robinson, U.S. National Zool. Park; W. A. Shear, Hampden- Sydney Coll.; G. W. Uetz, Univ. Cincinnati; C. E. Valerio, Univ. Costa Rica. The Journal of Arachnology (ISSN 0160-8202), a publication devoted to the study of Arachnida, is published three times each year by The American Arach- nological Society. Memberships (yearly): Membership is open to all those in- terested in Arachnida. Subscriptions to The Journal of Arachnology and American Arachnology (the newsletter), and annual meeting notices, are included with mem- bership in the Society. Regular, $30; Students, $20; Institutional, $80 (USA) or $90 (all other countries). Inquiries should be directed to the Membership Secretary (see below). Back Issues: Patricia Miller, P.O. Box 5354, Northwest Mississippi Community College, Senatobia, Mississippi 38668 USA. Telephone: (601) 562- 3382. Undelivered Issues: Allen Press, Inc., 1041 New Hampshire Street, P.O. Box 368, Lawrence, Kansas 66044 USA. THE AMERICAN ARACHNOLOGICAL SOCIETY PRESIDENT: Matthew H. Greenstone (1995-1997), Plant Science & Water Conservation Laboratory, USDA; Stillwater, Oklahoma 74075 USA. PRESIDENT-ELECT: Ann L. Rypstra (1995-1997), Dept, of Zoology, Miami University, Hamilton, Ohio 45011 USA. MEMBERSHIP SECRETARY: Norman I. Platnick (appointed), American Museum of Natural History, Central Park West at 79th St., New York, New York 10024 USA. TREASURER: Gail E. Stratton, Department of Biology, Rhodes College, Mem- phis, Tennessee 38112-1690 USA. BUSINESS MANAGER: Robert Suter, Dept, of Biology, Vassar College, Pough- keepsie, New York 12601 USA. SECRETARY: Alan Cady, Dept, of Zoology, Miami Univ., Middletown, Ohio 45042 USA. ARCHIVIST: Vincent D. Roth, Box 136, Portal, Arizona 85632 USA. DIRECTORS: James Carico (1995-1997), Pat Miller (1993-1996), Robert Su- ter (1995-1997). HONORARY MEMBERS: C. D. Dondale, W. J. Gertsch, H. W. Levi, A. F. Millidge, W. Whitcomb. Cover illustration: Posterior view of a female trapdoor spider, Cyclocosmia torreya. This spider has been found in Torreya State Park in the panhandle region of Florida, USA. Photo taken about 1950 by a well-known arachnologist, the late H. K. Wallace. Publication date; 29 May 1997 @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). 1997. The Journal of Arachnology 25:1-10 DISTRIBUTION, MOVEMENT, AND ACTIVITY PATTERNS OF AN INTERTIDAL WOLF SPIDER PARDOSA LAPIDICINA POPULATION (ARANEAE, LYCOSIDAE) Douglass H. Morse: Department of Ecology and Evolutionary Biology, Box G-W, Brown University, Providence, Rhode Island 02912 USA ABSTRACT. The wolf spider Pardosa lapidicina Emerton 1885 occupies cobble beaches above the tide line about Narragansett Bay, Rhode Island, USA, and migrates back and forth with the tides. To my knowledge this is the first explicit report of such behavior in a spider. The species is common, attaining densities of over 30 individuals/0.5 m of shoreline. The spiders are confined to the beach from April until November, and 33% or more of the population moves back and forth with diurnal tides during clear or warm weather. Individuals may migrate on one day but remain above the tide line on the next. Small numbers remain on the highest part of the beach through most or all of the winter; others retire into the adjacent coastal scrub. Spiders occupy both stretches of open beach and beach with fringing salt-marsh grass (Spartina aiterniflora) beds. Those in the latter habitat do not migrate through Spartina after it has reached high density in June, remaining confined to the upper reaches of the beach. Numbers of F. lapidicina on open stretches of the beach exceed those in Spartina areas, and they appear to experience higher mortality in the latter area. Spiders in the low intertidal move frequently and hunt actively (cursorial strategy), those above the tideline move only 0.1 times as often and sun-bask frequently (sit-and-wait strategy). Individuals hunting in the low intertidal may capture more than one prey per day, including Diptera, Collembola, and amphipods. Although spiders are primarily terrestrial, or sometimes occupants of the fresh water sur- face (BristO'We 1958; Levi 1967), several spe- cies frequent salt marshes (e.g.. Teal 1962; Dobel et al. 1990) and other intertidal habitats, where they may even withstand tidal inunda- tion (e.g., Bristowe 1923; Bames & Bames 1954; Roth & Brown 1976). A few even live permanently within the rocky intertidal zone, experiencing regular submersion, some for most of a tidal cycle (Hickman 1949; Lamoral 1968; McQueen & McLay 1983). Others re- treat to higher sites in the vegetation as the tide encroaches (Bristowe 1958). The wolf spider Pardosa lapidicina Emerton 1885 (Ly- cosidae), the subject of this paper, exploits the intertidal span of cobble beaches in Narragan- sett Bay, Rhode Island, USA, moving from above the high-tide line to the low as the tide recedes, and retreating in front of its return. To the best of my knowledge, this behavior has not previously been explicitly documented in a spider. Although not tolerating submer- sion like a few species that frequent intertidal areas, it occupies the marine-land interface (high beach, intertidal) almost exclusively during most of the year, using this remarkable behavior to exploit periodically available hab- itat. Here I document P. lapidicina'^ abun- dance, distribution, movements, periodicity and prey on a cobble beach in Narragansett Bay. I then compare these results with those of other lycosids and with other reports of spi- ders in the intertidal zone. In particular, this analysis permits me to evaluate the hunting strategies of Pardosa C.L. Koch 1848, vari- ously described as either sit-and-wait or cur- sorial (e.g., Bristowe 1958; Ford 1978), and rate of prey capture, frequently stated not to exceed one per day (Edgar 1970; Nyfeller & Benz 1988). Pardosa lapidicina is a dark-colored wolf spider of 6~9 mm length, the females some- what larger than the males (Kaston 1948). Mature adults weigh 30-70 mg or more, and large immatures in early September weigh 15-35 mg (Eason 1969; D.H. Morse, unpubl. data). Like other Pardosa (Vogel 1971; Low- rie 1973; Fujii 1974), they are small, cursorial, and nomadic. Members of the population de- scribed here, both males and females, are a uniform dull black. Voucher specimens of P. 1 2 THE JOURNAL OF ARACHNOLOGY Terrestrial ¥egetatioii Figure 1. — Diagram of part of study area with both open cobble beach and Spartina areas. Tran- sects denoted as a) high-tide census, Spartina area; b) high-tide census, cobble beach; and c) low-tide census, with 0.5 X 0.5 quadrats, on cobble beach. lapidicina have been deposited in the National Museum of Natural History, Smithsonian In- stitution. THE STUDY AREA Spiders were studied at the Haffenreffer Es- tate of Brown University, Bristol, Bristol County, Rhode Island, from September 1993- September 1995. The study area is a cobble beach on the west shore of Mt. Hope Bay, a partially sheltered eastern arm of Narragansett Bay. Tides range from 0.6-2. 1 m (x = 1.4 m), and the distance from the top (landward edge) of the beach only reached by severe storms to the lowest intertidal averages 23 m. Most cob- bles range from 10-30 cm in diameter, and larger stones and bedrock protrude in some places. At the lower intertidal levels, some bladder wrack seaweed Fucus vesiculosus, as well as a variety of encrusting organisms, grow on the larger stones. The four main study sites are cobble beach- es, 30-120 m long, punctuated by fringing salt-marsh grass Spartina altermflora that has invaded the beach in several places (Fig. 1). Observations and experiments were conduct- ed on three of the beaches, with all collecting confined to the fourth, to avoid any possible effects of resulting changes in population numbers on censuses or behavior. The inter- vening Spartina areas vary between 125-150 m in length. Additionally, two narrow corri- dors of cobble beach (5 m, 2 m) ran through the Spartina. The Spartina areas average 7.5 m in width, and are dense, except for the land- ward edges. There, stem densities in the upper 50 cm range between 0.25-0.5 X that of the center, which averages 348.3 ± 48.6 stems/ 0.25 {n = 10). Spartina ends about 10 m from the top of the beach and 5 m above the lowest tides (Fig. 1). A windrow of Spartina straw occurs high on the beach, usually 1-3 m from its upper edge. Above Spartina, this windrow often reaches 30 cm in height and 125 cm in width. Buildups are much less extensive on the open beach, even absent in the longest stretches. Bertness (1984) describes the study area in further detail. The land above the beach is covered by second-growth forest and scrub, with hackberry Celtis occidentalis, red oak Quercus rubra, and red cedar Juniperus vir- giniana under 20 m dominating the tree level, and bittersweet Celastrus scandens, greenbrier Smilax sp., and poison ivy Rhus radicans of- ten climbing into this canopy. Other than for the vines, ground cover is sparse. METHODS Transects.- — -Transects were laid out in sev- eral places on the main study sites and in Spartina-occupicd sites between them (Fig. 1). Spiders were counted in 0.5 m wide strips centered on those lines. All rocks in these transects were moved during a census, per- mitting an accurate count. Two types of cen- suses were ran. At a site on an open beach I counted spiders at low tide in 0.5 m X 0.5 m quadrats along a gradient from the top to the bottom of the beach. I ran this census weekly or biweekly over the study period, and it per- mitted me to assess both their numbers and distribution over the supratidal —low tidal gra- dient (Fig. Ic). I also counted spiders in a se- ries of 0.5 m wide transects at high tide every other week during the second year of the study. Three areas were chosen randomly on the open beach (Fig. lb) and three above Spartina regions(Fig. la). Six transects were run at randomly-selected sites in each of these areas, for a total of 18 transects on the open beach and 18 in the Spartina regions. Census methods followed those for the long-term transect described above, except that I merely counted the total in each strip, making no ef- fort to document the position of individuals in the gradient from the top of the beach to the MORSE— INTERTIDAL WOLF SPIDERS 3 water. These transects averaged 10 m or less in length, fluctuating with tidal height, and reached nearly down to the uppermost fringes of Spartina where it grew. Census at low tide.— Spiders were also pe- riodically counted near the low tide line, both directly between Spartina and the low-tide line and at the same height on open beaches. This area ranged between 2-5 m in width, de- pending on the daily height of the tide. During these censuses I kicked the rocks in the entire area being censused to flush spiders for count- ing. Efforts to calibrate this and the more time-consuming, hand-turning technique used in the transect studies indicated that “kick- sampling” yielded counts approximtely half those of hand-sampling. Movements,~To determine whether all in- dividuals migrated down the open beach, I conducted a mark-and-recapture (re-sight) test. At low tide on Day 1 individuals from a stretch of 40 m along the beach were dusted with powdered micronite dye, red in the mid- dle-low intertidal and yellow above the pre- vious high tide. On the following day individ- uals were censused at both levels by kick-sampling. The one-day interval between marking and censusing assured mixing of the individuals, since spiders in the intertidal had to retreat to the supratidal during the follow- ing high tide. Two high tides thus intervened between the marking and the census, but the interval was short enough to minimize effects of molt, mortality, and recruitment. Three such markings were conducted, but inclement weather on the day following marking pre- vented quantitative sampling on two of these occasions. To determine the timing of movements up and down the beach, white plastic strips 167 X 3 cm (length X width) were placed flush with the substrate, and movements of spiders across them recorded, along with their direc- tion, over entire tidal cycles. Three strips were used simultaneously, one per observer, for a total length of 5 m. I measured frequency of movement both in the low intertidal and high intertidal -suprati- dal areas of the open beach by observing focal individuals for periods of up to 30 min, since some moved only infrequently. To measure rates of movement by individuals going down and up the open beach, I timed spiders moving toward and away from the water for 10 min periods. I remained stationary during these pe- riods, measuring actual distances after the ob- servations. Movements of individuals splashed by surf were measured for shorter periods of 2. 0-2. 5 min, the time during which they moved at high velocities. Distribution in wiiiter.^-I used several methods to establish the presence and distri- bution of P. lapidicina in the forest above the beach. In September 1993, six plastic jars (9 cm diameter) were sunk into the earth flush with the surface in the scrub woodland 3 m above the beach. All were placed equidistantly along the periphery of the first beach, main- tained through the fall, and their contents col- lected weekly. The jars were partly filled with leaves and litter to minimize possible preda- tion and permit subsequent release. Litter above the beach was also searched for spiders between September-November 1993. Ten of leaf litter were turned with a coarse rake each month between May-November 1994, both at 3 m and 5-10 m above the beach. Activity and prey capture.— Focal obser- vations of spiders on the beach permitted compilation of activity patterns and time bud- gets. Prey capture was recorded when noted, and 10 min observation periods of individuals permitted quantification of the frequency of attacks and captures. RESULTS Population density and size.— The week- ly/biweekly transect census taken at low tide over the entire study period provided a com- parison among seasons. The maximum count of spiders in this 0.5 m wide transect during late Autumn 1993 was 37 (Fig. 2). If repre- sentative of this 32 m long beach, over 2000 individuals entered winter in this one area alone. In Spring 1994, I recorded a maximum of 20 spiders, suggesting a 45% loss of indi- viduals over the severe winter of 1993-94. Autumn counts in 1994 were only roughly half those of 1993, and those of 1995 were intermediate between those of 1993 and 1994 (Fig. 2). The cohort hatched in Spring-Sum- mer 1994 never attained the densities of the 1993 cohort during Autumn 1993; however, their maximum numbers in Spring 1995 ex- ceeded those from the preceding autumn (Fig. 2), suggesting low winter mortality and colo- nization from adjacent areas. Numbers of adults decreased rapidly during 4 THE JOURNAL OF ARACHNOLOGY Figure 2. — Total numbers of spiders counted at low tide each 1-2 week period in 0.5 m wide transect on open beach, and numbers captured in pitfall traps. the summer, coinciding with the appearance of young (Fig. 2), strongly suggesting that few if any individuals survived more than one year. The last adults were seen on this transect on 17 August 1994 and 1 August 1995, although occasional individuals were subsequently seen elsewhere in the study area on later dates, the last being two adults on 10 October 1994. Spatial distribution. ““Numbers of both adults and juveniles on bare cobble sites ex^ ceeded those on Spartina sites (Fig. 3, F = 0.01 in a binomial test). During spring and summer, adults at bare cobble sites exceeded those at Spartina sites by 40% in 1994, and nearly two-fold in 1995. Differences were considerably smaller among early juveniles. Figure 3. — Censuses at high tide in 0.5 m wide transects on open beaches and in Spartina areas. MORSE— INTERTIDAL WOLF SPIDERS 5 Table L — Numbers of Pardosa at low tide on shoreline below Spartina and at same height on open beach. Habitat Num- ber counts Num- ber spiders Length of shore (m) Spi- ders/m Below Spartina 21 0 235 0 Low open beach Narrow breaks 8 175 295 0.6 in Spartina 8 38 40 1.0 with numbers on the open beach averaging only 10% higher than those above Spartina through September 1994, before strongly di- verging in late September. A similar pattern occurred in 1995, although numbers diverged by early September (Fig. 3). Spartina growth generated an absolute bar- rier to the movement of spiders into the low intertidal area during most of the season. In six, 0.5 m wide transects made down the beach through Spartina in late summer, no spiders were found between the landward edge of Spartina and the low water line. Spi- ders migrating down the adjacent bare cobble beach moved no more than 8 m laterally onto the rocks below the Spartina fringe. At the landward edge of the Spartina, spiders pene- trated no more than 40 cm into the vegetation. During summer and autumn I never found spi- ders in the low rocky areas immediately below Spartina, but found them common near the low-tide line on adjacent open cobble beaches (Table 1). They also readily moved through narrow corridors in Spartina into the low in- tertidal in densities comparable to or greater than those of the wider beaches (Table 1). Seasonal change.-— I did not find individ- uals in the low intertidal area after 16 October or before 17 April in the transect census, al- though recording them in the low intertidal during other field work as early as 25 March and as late as 20 October. Thus, the spiders confined movement into the lower reaches of the intertidal to the warmer part of the year. Even then (17 April- 16 October), significant- ly more spiders (> 2.5 X) occupied the high beach than the area below the wrack line in the transect (x^ 38.9, = 1, P < 0.001 in a one- sample test). In contrast, spiders moved over Spartina turf before new grass sprouted in the spring, continuing while shoots were sparse and only a few cm tall (late March-late April). Num- bers on the beach below Spartina reached 35/100 m at such times. As the grass grew taller and denser in May, only occasional spi- ders penetrated it (< 1 vs. 20-65 individuals/ 100 m near the low-tide line on the open beach). No spiders were seen below Spartina after early June. The movement of spiders into the adjacent forest during late fall was sudden and marked (Fig. 2). In 1993, I searched weekly for indi- viduals under stones and in the litter within a 5 m strip above the beach, as well as moni- toring six pitfall traps located 3 m above the beach. Neither the searches nor the pitfall traps yielded any spiders until 14 November 1993, when 15 individuals were captured in the pitfall traps (Fig. 2), and other individuals were found under rocks and in the vicinity of the traps. Several more individuals were cap- tured in the traps over the following two weeks, and then captures declined to only 1- 2/week, with the last individuals captured on 12 December 1993 (Fig. 2). In monthly searches of litter from May-November 1994 (0.1-10 m above beach), I found no individ- uals until 6 November, when I located three. Numbers of individuals on the beach de- clined markedly on the first week in 1993 that spiders were captured in the pitfall traps, and in subsequent weeks only a few individuals were recorded on the beach (0-4). However, spiders occupied this site until snow and ice completely covered it on 5-6 January (Fig. 2). I also found Pardosa at this site on 12 March (Fig. 2), shortly after snow and ice had melted from the upper edge of the beach. Numbers of individuals on the main transect also de- clined during the snowless 1994-95 winter, although seldom to the level of 1993-94 (Fig. 2). However, they largely disappeared from the replicated transects (Fig. 3), suggesting that activity on the beach was confined to a few sites during the middle of the winter. Daily activity. — At low tide spiders ranged over the entire vertical expanse of the open cobble during the day in dry, warm weather, although large numbers, usually a majority, occupied the supratidal part. Numbers of in- dividuals at or above the neap high-tide line (the level at which algal wrack accumulated), about 5 m below the rock-forest interface, ex- 6 THE JOURNAL OF ARACHNOLOGY Distance from top of beach (m) Figure 4. — Distribution of spiders at or near low tide from landward edge of beach to water. Cu- mulative results of 0.5 X 0.5 quadrats pooled into 1 m bars. Low-tide line variable, but 15 m or more. ceeded those below it, cumulatively over four- fold (Fig. 4: G = 103.6, P < 0.001 in a G- test, df = 1, for adults and subadults; G = 29.1, P < 0.001 in a G-test, df = 1, for ju- veniles), although the area below the high-tide line was often three or more times greater than the upper area. Spiders sheltered under rocks during peri- ods of rain and heavy overcast. After record- ing no individuals at the surface on three oc- casions (September-October 1993), I confined fieldwork to favorable weather. On two night visits during warm weather (> 15 °C), all spi- ders were also sheltered under rocks. Not all individuals selected the same levels of open cobble beach on subsequent days (Ta- ble 2), although significantly more were re- sighted at their original marking level than predicted by chance (Z = 1.833, one-tailed bi- nomial test, P < 0.05 for above the high-tide line, Z = 1.658, one-tailed binomial test, P < 0.05 for the intertidal). The two groups did not differ in their tendency to shift from one level to the other on the following morning (G = 0.08, df = 1, P > 0.7 in G-test), or in rate of recapture (G = 2.06, df = 1, P > 0.1 in G- test). Individuals carrying egg sacs or young did not venture into the lower tidal reaches: I have yet to record such a spider over 5 m below the high-tide line. Movement and activity.— Considerable numbers of individuals moved up and down the beach, the juveniles beginning in their third instars. During a representative mid-day, low-tide episode on 19 July 1994, a minimum of 26 adults and 69 young moved down and Table 2. — Results of mark-resight test of spiders captured and marked in supratidal and low-middle intertidal areas during low tide in May 1995, Re- sightings made one day after marking. Site Number marked Resight- ing same color Resight- ing opposite color Sight- ing un- marked Supratidal Low & mid- 120 24 12 64 intertidal 113 28 16 71 back over a 5 m wide stretch 12.5 m below the upper edge of the beach. Extrapolated to the 120 m of this beach, over 600 adults and 1650 young moved from the high tidal to the low tidal area on that day. Spiders moved al- most constantly across the counting strips, but the greatest numbers lagged the outgoing tide considerably, preceding the low tide by only 30-45 min (Fig. 5). Several even crossed downward after low tide, and a trickle of in- dividuals continued moving downward for an- other 2.25 h (Fig. 5). Thus, a majority of their time in the low intertidal was spent as the tide returned toward its mid-point. Although sev- eral spiders returned only shortly before water inundated the census strip, the largest numbers preceded the resurging tide by 45-60 min (Fig. 5). Adults and young did not clearly dif- fer in times of movement. Movement down the beach at low tide pro- ceeded at 17 m/h (2.8 ±1.9 m/10 min, n = 10), from below the wrack line to within 2 m of the water line. Return up the beach was more rapid: spiders within 5 m of the edge of the water on an advancing tide moved 42 m/h {n = 11) (7.0 ± 4.2 m/10 min), a rate that would take them the entire breadth of the beach in under one hour. Eight individuals splashed by surf moved even faster for short periods after this stimulus (2.0-2. 5 min), cov- ering 2-3 m (70.9 ± 7.8 m/hr) over this short period. This movement consisted of several consecutive runs of up to 1 m, punctuated by rests of only a few seconds. The low variance of this sample (extremes — extrapolated rates of 60-75 m/hr) suggests that these spiders had approached their maximum possible speeds. In contrast, most upward movements of un- disturbed spiders ranged between 5-50 cm (15.2 ± 8.2 cm, n = \1 individuals and 182 MORSE— INTERTIDAL WOLF SPIDERS 7 Figure 5. — Movements by spiders up and down tide line over a 5 m white plastic barrier placed 12.5 m below the upper edge of beach (mid-tide) on a routine day, 19 July 1994. Time in hours of day, period denoted as “dry” (< 0700~“>1500) refers to time the barrier region was not covered with water. Adults and juveniles combined. moves), only two moves exceeding this dis- tance (75, 60 cm). Spiders were significantly more active in the low intertidal than above the tide line, even when they were not migrating. Move- ments in the low intertidal area {n = 11) oc- curred over 10 times as frequently as those in the high intertidal and supratidal {n = 10) at the same time (every 29.2 ± 37.1 sec vs. ev- ery 412.8 ± 624.9 sec; P < 0.001 in a two- tailed Mann Whitney C/-test), a reflection of the individuals in the low intertidal being con- stantly on the move, often hunting, while those in the high intertidal were largely sun- basking. Spiders usually avoided direct contact with the water, but regularly reached the water’s edge, although most frequently stopping a minimum of 0.5 m from it. Nine individuals washed into the water by unusually high waves curled their legs under them, and all Table 3. — Prey attacked and captured by Pardosa on low beach, June-August (21.17 h observations). * Spider snapped in same way as it attacked prey, but object not seen. Prey Attacked Captured Captures//h Diptera 6 4 0.19 Collembola 10 7 0.33 Others 8 2 0.09 Not seen* Total 10 0.52+ eventually managed to crawl out onto a near- by stone and retreat rapidly upshore {P — 0.004 in a two-tailed Binomial Test). I never saw any suggestion that these spiders re- mained submerged during a tidal cycle. Prey. — I observed 34 attacks or captures in the low intertidal, an average of 1.6/h (Table 3). Spiders attacked small flies feeding in the low intertidal (seaweed flies (Coelopa frigida, Coelopidae)), Collembola (Anurida maritima, Hypogastruridae), and unknowns, many of which probably were flies. Additionally, sev- eral apparent strikes were noted, for which the target was not seen. Probably most of these strikes were directed at Collembola (see Table 3). A majority of observed attacks directed at both the flies and Collembola was successful. Other untimed observations on hunting and prey capture resembled these. Additionally, four captures of newly-molted amphipods (Orchestia sp.) warrant note. Amphipods are abundant under the rocks of the intertidal (scud, Jassa sp.; Gammarus sp.) and near the wrack line (beach fleas, Orchestia sp.). DISCUSSION Intertidal area.— The tide-punctuated mi- gratory movement of P. lapidicina is highly unusual (or unreported), I have not found a similar pattern in the literature for any spider, although Lamoral (1968) reported that the permanently intertidal Desis formidabilis (O.P. -Cambridge 1890) (Desidae) from South Africa exhibited a strong sense of tidal 8 THE JOURNAL OF ARACHNOLOGY rhythm, remaining in their nests when tides were high at night, their normal period of ac- tivity. However, P. lapidicina's behavior does somewhat resemble that of P. pullata (Clerck 1757), a European species Bristowe (1923, 1958). Bristowe (1958) reported them “lying idly on the pebbles piled up by the sea”, but did not indicate whether they routinely moved down into the intertidal as P. lapidicina does. However, he noted that these P. pullata re- treated up the tide line into vegetation when exceptionally high tides occurred, but were occasionally trapped by these tides. If so trapped, they curled their legs under them, as does P. lapidicina, and floated passively on the wave until it receded, depositing them on the pebble substrate. Then they ran toward the land before the next wave arrived, and simi- larly to P. lapidicina, always succeeded in es- caping. Pardosa lapidicina is common enough to play an important energetic role in the inter- tidal zone. Such a role would not be unusual, since spiders, especially lycosids, are the dominant invertebrate predators in some streamside (Vlijm et al. 1963), salt-marsh (Schaefer 1974), grassland (Van Hook 1971) and forest habitats (Moulder & Reichle 1972). Effect of Spartina barriers. — Spiders of the open beach appeared to be more success- ful than those in the Spartina areas. Densities were almost always higher on the open beach, although differences were initially small in the immature cohort during mid-summer. How- ever, the rapid decrease in numbers of spiders in the Spartina areas during the fall suggests lower survival there than on the open beaches. Spiders in the Spartina areas may have been in poorer energetic condition than those from the open beach. Activity. — The behavior of individuals in supratidal and intertidal areas differed mark- edly. In the supratidal area, spiders moved rel- atively infrequently and often sunbasked. Their behavior resembled that of several other Pardosa species, which employ sit-and-wait behavior almost exclusively, waiting until prey come very close to them (Edgar 1969; Kronk & Riechert 1979; Nakamura 1982) or even touch them (Fuji! 1974; Ford 1978), only then rapidly attacking. Traditionally, lycosids, including Pardosa, have been considered cur- sorial predators that tracked down their prey (e.g., Comstock 1940; Bristowe 1958), and the previously-noted authors go to consider- able ends to “correct” the record. Ford (1978) noted that his P. amentata (Clerck 1757) spent no more than 278 sec/day moving (0.0032%/ day). The P. lapidicina in the supratidal and high intertidal areas exhibited a sit-and-wait pattern closely resembling the one described by the more recent workers, which simulta- neously allowed them to sun-bask on clear days. Those on the low beach, however, moved much more frequently, often stalking and leaping at flies, active behavior similar to that described by Bristowe and other earlier workers. In light of these striking differences and the disagreement in the literature, detailed time budgets of a representative range of spe- cies are needed. The high rates of movement observed in wave-splashed spiders resembled the maxi- mum movement rates of both P. lugubris (Walckenaer 1802) and Xerolycosa nemoralis (Westring 1861) continually chased by Bris- towe (1939), who found that they could run a maximum of 1.8 and 1.5 m, respectively, be- fore experiencing temporary exhaustion. These figures appear comparable to the rapid moves of 2-3 m seen in bursts of up to 1 m by P. lapidicina not subjected to these artifi- cial stimuli. Hunting. — Studies of lycosids in the field reveal very low percentages of individuals found feeding on prey (2% -Nakamura 1982; 6% -Nyffeler & Benz 1981a), consistent with a low intake rate. Some lycosids are estimated to capture no more than one prey per day in the field (Edgar 1970; Nyffeler & Benz 1988), although they routinely accept several per day in the laboratory (Miyashita 1968; Samu 1993). Observations in this study suggested that individuals in the low intertidal area rou- tinely captured more than one prey per day; however, many of those prey were tiny col- lembolans, which provide only a minute store of resources. Although amphipods are abun- dant on the beach, they typically have hard carapaces and hence may be difficult to pro- cure (Moulder & Reichle 1972), unless they have recently molted. All amphipod prey in this study had recently molted. Other small lycosids also take small prey (Fitch 1963; Dondale et al. 1972; Nyffeler & Benz 1981a, 1981b) and experience considerable difficulty MORSE— INTERTIDAL WOLF SPIDERS 9 taking any with a hard carapace (Moulder & Reichle 1972). ACKNOWLEDGMENTS I thank J. Blumenstiel, A. Choi, C. Harley, and M. Weiss for assistance in the field, R.L. Edwards for identifying Pardosa lapidicina and other spiders, and C. Harley for comments on the manuscript. LITERATURE CITED Bames, B.M. & R.D. Barnes. 1954. The ecology of the spiders of maritime drift lines. Ecology, 35:25-35. Bertness, M.D. 1984. Habitat and community modification by an introduced herbivorous snail. Ecology, 65:370-381. Bristowe, W.S. 1923. A British semi-marine spider. Ann. Mag. Nat. Hist., Ser. 9:154-156. Bristowe, W.S. 1939. The comity of spiders. Ray Society, London. Bristowe, W.S. 1958. The world of spiders. Col- lins, London. Comstock, J.H. 1940. The spider book. (W.J. Gertsch, ed.). Comstock, Ithaca, New York. Dobel, H.G., R.E Denno & J.A. Coddington. 1990. Spider (Araneae) community structure in an in- tertidal salt marsh: effects of vegetation structure and tidal flooding. Environ. EntomoL, 19:1356- 1370. Dondale, C.D., J.H. Redner & R.B. Semple. 1972. Diel activity periodicities in meadow arthropods. Canadian J. Zool., 50:1155-1163. Eason, R.R. 1969. Life history and behavior of Pardosa lapidicina Emerton (Araneae: Lycosi- dae) J. Kansas EntomoL Soc., 42:339-360. Edgar, W.D. 1969. Prey and predators of the wolf spider Pardosa lugubris. J. Zool. (London), 159: 405-411. Edgar, W.D. 1970. Prey and feeding behaviour of adult females of the wolf spider Pardosa amen- tata (Clerck). Netherlands J. Zool., 20:487-491. Fitch, H.S. 1963. Spiders of the University of Kan- sas Natural History Reservation and Rockefeller Experimental Tract. Univ. Kansas Mus. Nat. Hist. Misc. Publ., 33:1-202. Ford, M.J. 1978. Locomotory activity and the pre- dation strategy of the wolf- spider Pardosa amen- tata (Clerck) (Lycosidae). Anim. Behav., 26:31- 35. Fujii, Y. 1974. Hunting behaviour of the wolf spi- der, Pardosa T-insignita (Boes. et. Str.). Bull. Nippon Dental Coll., Gen. Educ., 3:134-148. Hickman, V.V. 1949. Tasmanian littoral spiders with notes on their respiratory systems, habits and taxonomy. Pap. Proc. Roy. Soc. Tasmania, 1948:31-43. Kaston, B.J. 1948. The spiders of Connecticut. Connecticut State Geol. Nat. Hist. Surv. Bull. 70. 874 pp. Kronk, A.E. & S.E. Riechert. 1979. Parameters af- fecting the habitat choice of a desert wolf spider Lycosa santrita Chamberlin and Ivie. J. Arach- nol.,7: 155-166. Lamoral, B.H. 1968. On the ecology and habitat adaptations of two intertidal spiders, Desis for= midabilis (O.P. Cambridge) and Amaurobioides africanus Hewitt, at “The Island” (Kommetjie, Cape Peninsula), with notes on the occurrence of two other spiders. Ann. Natal Mus., 20:151-193. Levi, H.W. 1967. Adaptations of respiratory sys- tems of spiders. Evolution, 21:571-583. Lowrie, D.C. 1973. The microhabitats of western wolf spiders of the genus Pardosa. Ent. News, 84:103-116. McQueen, D.J. & C.L. McLay. 1983. How does the intertidal spider Desis marina (Hector) re- main under water for such a long time? New Zealand J. Zool., 10:383-392. Miyashita, K. 1968. Quantitative feeding biology of Lycosa T-insignita Boes et Str. (Araneae: Ly- cosidae). Bull. Nat. Inst. Agric. Sci. (Japan) C, 22:329-344. Moulder, B.C. & D.E. Reichle. 1972. Significance of spider predation in the energy dynamics of forest-floor arthropod communities. Ecol. Mon- ogr., 42:473-498. Nakamura, K. 1982. Prey capture tactics of spi- ders: an analysis based on a simulation for spi- der’s growth. Res. Popul. Ecol., 24:302-317. Nyffeler, M. & G. Benz. 1981a. Einige Beobach- tungen zur Nahrungsdkologie der Wolfspinne Pardosa lugubris (Walck.), Deutsches Ent. Z., 28:297-300. Nyffeler, M. & G. Benz. 1981b. Freilanduntersu- chungen zur Nahrungsokologie der Spinnen: Beobachtungen aus der Region Zurich. Anz. Schadlingskde. Pflanzenschutz, Umweltschutz, 54:33-39. Nyffeler, M. & G. Benz. 1988. Feeding ecology and predatory importance of wolf spiders {Par- dosa spp.) (Araneae, Lycosidae) in winter wheat fields. J. Appl. EntomoL, 106: 123-134. Roth, V.G. & W.L. Brown. 1976. Other intertidal air-breathing arthropods. Pp. 119-150, In Marine insects. (L. Cheng, ed.). North Holland Publ. Co. Samu, F. 1993. Wolf spider feeding strategies: op- timality of prey consumption in Pardosa horten- sis. Oecologia, 94:139-145. Schaefer, M. 1974. Experimental studies on the im- portance of interspecies competition for the ly- cosid spiders in a salt marsh. Proc. Int. Arachnol. Congr., 6:86-90. Teal, J.M. 1962. Energy flow in the salt marsh eco- system of Georgia. Ecology, 43:614-624. Van Hook, R.L, Jr. 1971. Energy and nutrient dy- 10 THE JOURNAL OF ARACHNOLOGY namics of spider and oithopteran populations in a grassland ecosystem. Ecol. Monogr., 41:1-26. Vlijm, L., A. Kessler & C.J.J. Richter. 1963. The life history of Pardosa amentata (Cl) (Araneae, Lycosidae). Entomol. Bericht., 23:75-80. Vogel, B.R. 1971. Individual interactions of Par- dosa. Armadillo Pap., 5:1-13. Manuscript received 13 November 1995, revised 1 September 1996. 1997. The Journal of Arachnology 25:11-19 NOTES ON THE REPRODUCTIVE BIOLOGY AND SOCIAL BEHAVIOR OF TWO SYMPATRIC SPECIES OF PHILOPONELLA (ARANEAE, ULOBORIDAE) Deborah R. Smith: Department of Entomology, Haworth Hall, University of Kansas, Lawrence Kansas 66045 USA ABSTRACT. Populations of the facultatively communal species Philoponella oweni (Chamberlin 1924) and Philoponella arizonica (Gertsch 1936) (Uloboridae) occur sympatrically in the Chiricahua mountains of southeastern Arizona. This study compares reproductive biology, structure of communal groups, and feeding rates of the two species, and documents differences in their phenology, webs, web construction sites, egg-cases and spiderlings. I suggest environmental factors that may select for different reproductive strategies in the two species. Many members of the spider family Ulo- boridae have been observed living in groups (Opell 1979; Muma & Gertsch 1964). For only a few of these species has the nature of their group-living behavior been investigated: the facultatively communal Philoponella ow- eni (see Smith 1982, 1983) and P. semiplu- mosa (Simon 1893)(see Lahmann & Eberhard 1979); P. republicana (Simon 1891), with its large, semi-permanent colonies (Smith 1985; Binford & Rypstra 1992); and a west African Philoponella Mello-Leitao which was ob- served in a very large colony (Breitwisch 1989). This study compares reproductive biology, feeding rates and group-living behavior of two sympatric populations of group-living Philo- ponella, P. oweni (Chamberlin) and P. arizon- ica (Gertsch) (Uloboridae). Notes on the nat- ural history of the two species are also presented, including structure of the egg- cases, the structure of the webs and substrates used for web-building in the two species. Populations of Philoponella oweni and P. arizonica are broadly sympatric in the south- western United States and northeastern Mex- ico (Muma & Gertsch 1964; Opell 1979). The basic life cycles of the two species are similar in southeastern Arizona. Field observations indicate that both are annual; sub-adults emerge from overwintering sites in the spring (early April to early June, depending on ele- vation). Mating takes place in late spring and early summer. In general, males are shorter- lived than females and disappear from the population during the course of the summer. Females can lay eggs throughout the summer and may survive until early autumn, but in the populations studied no adult females overwin- tered for a second breeding season. Immatures hatch and emerge from the egg- case during the summer. As is true of other uloborid spiders, the young can spin webs in the first post-emergence instar (Szlep 1961; Eberhard 1977). The newly emerged spider- lings lack a functional cribellum; and the orb webs they produce are distinctive, containing many hundreds of radial threads without a sticky spiral. Later instars possess a functional cribellum and produce webs that are similar or identical to those of adults in form (Szlep 1961; Eberhard 1977). The young remain with the female for a variable length of time and attach their orbs to her web; during the course of the summer some or all of them disperse out of the ma- ternal web and build independent webs (Smith 1982). Young spiders over- winter as sub- adults or younger immatures, and emerge the following spring to form the next generation of reproductives. Both species are facultatively communal; that is, adult females of both species can be found living in small communal groups or as solitary individuals (Smith 1982, 1983). In the communal groups each female constructs her own orb web and defends it against other adult females. The orbs are joined by their support lines and space webbing. 11 12 THE JOURNAL OF ARACHNOLOGY Figure 1. — Egg-cases of Philoponella oweni (above) and Philoponella arizonica (below). Both species suffer from egg-parasitism by the chalcidoid wasp Arachnopteromalus dasys Gordh (Pteromalidae) (Gordh 1976), which greatly affects reproductive success of indi- vidual females. The female wasp oviposits in a uloborid spider’s egg-case, and the larvae consume the contents of the spider’s eggs, leaving behind empty spider egg shells. The wasp larvae then pupate inside the spider egg- case and emerge as adults. If an egg-case is parasitized, all of the spider eggs inside are killed (Smith 1982). METHODS Observations and collections were made at several sites in the Chiricahua Mountains in southeastern Arizona: the Southwestern Re- search Station of the American Museum of Natural History; South Fork Canyon, Cave Creek Canyon, and Herb Martyr Reservoir in the Coronado National Forest; and the town of Portal (Cochise County). Population censuses: Adult females and some males were individually color-marked with dots of fast-drying enamel paints (Tes- ter’s model airplane paint) and censused 2-5 times per week. From 25 April--20 August 1977 I censused a marked population of P. oweni in Cave Creek Canyon at an elevation of approximately 1695 m. This population was destroyed by a flood in 1978. From 3 June- 18 September 1979 I censused marked populations of P. oweni and P. arizonica in South Fork Canyon. The P. arizonica popu- lation was located in the lower part of the can- yon (1525-1660 m) while the P. oweni pop- ulation was in the upper part of the canyon (1630-1730 m). At each census visit I noted the presence of adult females, males, imma- tures or egg-cases. Reproductive biology: When a female pro- duced an egg-case it was given a color mark corresponding to that of the mother. To deter- mine mean clutch size (number of eggs per egg-case) for each species I collected the egg- cases after the young had emerged and examined their contents with a dissecting mi- croscope. As is true of most spiders, the young of Philoponella do not emerge from the egg-case immediately upon hatching from the egg; they remain in the egg-case for one in- star, molt, and emerge as second instar spi- derlings leaving behind both empty egg-shells and cast-off exoskeletons. In healthy egg-cases I used the number of egg shells as a measure of the number of eggs laid in the egg-case (Smith 1982). A parasitized egg-case can be recognized by the absence of exoskel- etons from first instar spiderlings and presence of wasp pupal skins. Although all the spider eggs are killed in a parasitized egg-case, the egg-shells are still visible and were used to infer the original clutch size. Colony structure: I noted the size and or- ganization of all communal groups formed by each species in several locations: P. oweni in Cave Creek Canyon (1977), South Fork Can- yon (1979, 1980) and Herb Martyr (1980); P. arizonica in South Fork Canyon (1979, 1980). Feeding rates: From 21-27 August 1979, I compared feeding rates of solitary and com- munal females of the two species in South Fork Canyon using a trapline census method. Censuses were carried out from 0600-1800 h. Daily census periods were 4-6 h long, for a total of 29 h of observation. Every hour I vis- ited each female in the study area and record- ed whether or not she was feeding or engaged in prey capture. Web structure: In July and August of 1980 I measured the webs of all adult females found in the South Fork Canyon study area: 22 webs of P. oweni and 19 webs of P. ari- zonica. I measured longest diameter of the orb webs and deviation (to the nearest 10°) of the plane of the orb webs from horizontal, and counted number of radii and number of spiral turns along the longest radius. I also noted the amount and position of “space webbing” (ir- regular tangles of threads), the presence or ab- sence of a stabilimentum and the general ap- pearance of the webs. Habitat: To evaluate the distribution of the two species with respect to elevation I located SMITH— NOTES ON PHILOPONELLA 13 Table L — Phenological data for Philoponella oweni and Philoponeila arizonica. * Still active when field observations ended. oweni (1977) Cave Creek oweni (1979) South Fork arizonica (1979) South Fork Date census began 26 April 3 June 3 June First adult female seen 5 May 3 June 3 July First adult male seen 5 May 3 June 3 July First egg-case seen 16 June 12 June 3 July First hatchlings seen 28 June 2 July 5 July Last adult male 4 July 27 June 21 July Last adult female 20 August* 1 1 July 18 September* End of census 20 August 18 September 18 September all collection sites (eight for P. oweni, six for P. arizonica) on a topographical map. I also recorded the substrates used for web attach- ment by members of each species. In July and August 1977 I recorded the substrates used for 25 P. oweni webs in Cave Creek Canyon, and in August 1979 I recorded the substrates of 36 P. oweni and 31 P. arizonica webs in South Fork Canyon (in each case, this represented all adult webs present in the study sites). Egg-cases: In July 1980 I collected 24 emp- ty P. oweni egg-cases and 33 empty P. ari- zonica egg-cases and measured maximum width and length to the nearest mm using dial calipers, and noted their color, shape and or- namentation. RESULTS Population censuses: Table 1 gives the dates of first sightings of age and sex classes of both species. These data are limited by the starting and finishing dates of the censuses, but still show differences between adjacent populations of P. oweni and arizonica, P. ow- eni adults appear sooner (at a given altitude) and adult P. oweni males disappear from these populations by late June and early July. Adults of P. arizonica appear later, and the adult males persist in populations until late July. There was little temporal overlap between P. oweni males and P. arizonica females in ad- jacent populations. Reproductive biology: Table 2 presents data on reproductive parameters for all females in the study areas, whether or not complete re- cords of their reproductive history could be made. In 1979 I had complete reproductive histories for 31 P. oweni females and 25 P. arizonica females. These data are presented in Table 3. Both tables show that P. oweni fe- males produce fewer egg-cases per female and lay more eggs per egg-case than do P. arizon- ica females. The behavior of females with egg-cases also differs between the two spe- cies. A female P. oweni about to construct an egg-case leaves her prey-capture orb and moves to the retreat area, a protected area near the orb usually under a rock or log, and con- structs the egg-case there. There she remains, holding the egg-case and (presumably feeding little or not at all) until the young emerge, a period of approximately 20 days. After the young emerge the mother discards the egg- case, leaves the retreat, and spins a new prey capture-orb. P. oweni females usually have one or at most two egg-cases at a time. The time interval between successive egg-cases produced by a female is more than a week, typically 2-3 weeks. In contrast, females of P. arizonica were never seen to leave the orb with their egg- cases. These females suspend their spindle- shaped egg-cases from the hub of their hori- zontal orbs. As new egg-cases are constructed, at intervals of 4-10 days, they are attached to the egg-cases already hanging in the web to form a long, slender stick (Fig. 1). The female continues feeding while the eggs and young mature. P. arizonica females sometimes have as many as 8 egg-cases in the web at once. Egg-case parasitism by the wasp A. dasys (Fig. 2) is a major source of mortality in both species. In general, a higher proportion of the egg-cases of P. arizonica than of P. oweni are attacked by egg-case parasites. In 1979, 14% of the egg-cases produced by P. oweni females and 27% of the egg-cases pro- 14 THE JOURNAL OF ARACHNOLOGY Table 2. — Reproductive parameters for Philoponella oweni and Philoponella arizonica. Cave Creek Canyon in 1977 and South Fork Canyon in 1979. Statistical tests are for differences between adjacent Philoponella oweni and Philoponella arizonica populations in South Fork Canyon in 1979. = Mann Whitney f/-test; = two-tailed Mest for samples with equal variance, t = 3.21, 1 df; x^ = 7 .10, I df; \ Mann Whitney f/-test. Year Mean SD Range n P Egg-cases per female oweni 1977 1.29 0.50 1-3 49 females oweni 1979 1.5 10.8 1-4 44 <0.00T arizonica 1979 3.2 2.0 1-8 50 Clutch size oweni 1977 35.2 15.6 10-85 51 egg-cases oweni 1979 50.0 17.0 14-89 50 <0.001*’ arizonica 1979 21.8 8.9 6-49 112 Egg-cases parasitized oweni 1977 23.5% 51 egg-cases oweni 1979 14.0% 50 0.073^^ arizonica 1979 27.0% 112 Females with > 1 oweni 1977 40.0% 30 females egg-case parasitized oweni 1979 22.0% 32 <0.007^ arizonica 1979 55.0% 27 Live young per egg-case oweni 1977 27.1 20.1 0-79 51 egg-cases oweni 1979 42.5 21.5 0-88 50 <0.00L arizonica 1979 15.8 12.0 0-41 112 duced by P. arizonica females were parasit- ized (this difference is not significant; = 3.21, 1 df, Table 2). For females for whom complete reproductive histories were recorded (Table 3), significantly more egg-cases of P. arizonica than of P. oweni were parasitized. Similarly, for all females and for females with complete reproductive records, a higher pro- portion of P. arizonica females than P. oweni females lost at least one egg-case to parasites. Because the average clutch size of P. oweni females is larger than that of P. arizonica fe- males, the appropriate comparison of repro- ductive effort and reproductive success is life- time egg and spiderling production of individual females (Table 3). Mean lifetime egg production by P. oweni and arizonica fe- males did not differ significantly, nor did mean lifetime production of live spiderlings (Mann Whitney f/-test). However, 19% of P. oweni females (6 of 31) lost all of their eggs to parasites, while only 4% of the P. arizonica females (1 of 25) was similarly affected. While this difference is not significant (x^ = 2.9, 1 df), it does suggest that P. arizonica's habit of packaging lifetime egg production into many small clutches may reduce the risk of losing an entire lifetime of egg production to parasites. A comparison can also be made between Table 3. — Lifetime reproductive parameters for Philoponella oweni and P. arizonica for whom complete life histories are known (South Fork Canyon, 1979). “ = Mann Whitney f/-test; ^ = two-tailed f-test for samples with equal variance, t = -1.73; x^ ^ 18.16, 1 df, x^ = 8.16, 1 df; Mann Whitney U-test. Mean SD Median Range n P Egg-cases per female oweni 1.3 0.5 1.0 1-2 31 females <0.00T arizonica 3.9 1.9 4.0 1-8 25 Total eggs oweni 66.3 21.4 68.0 30-100 31 females 0.09*’ (ns) arizonica 82.4 42.3 76.0 17-208 25 Egg-cases parasitized oweni 15.0% 40 egg-cases 0.00002^^ arizonica 54.0% 97 Females with > 1 oweni 16% 31 females 0.0043^* egg-case parasitized arizonica 52% 25 Total young oweni 54.8 32.7 62.0 0-100 31 females 0.9L(ns) arizonica 58.0 36.2 55.0 0-159 25 SMITH— NOTES ON PHILOPONELLA 15 Figure 2. — Arachnopteromalus dasys on Philo- ponella oweni egg-cases. the solitary and group-living members of each species. It was reported earlier (Smith 1982) that on average, communal females P. oweni produced more eggs per egg case than solitary females, though they did not differ in mean number of egg cases per female (1977: soli- tary females, 26.9 SD ±13 eggs per egg case, n = 13 cases; communal females, 37.3 ±15.1 eggs per egg case, n — 38 egg cases, t = 2.17, P < 0.05. 1979: solitary females 44.3 ± 15.3 eggs, n = 26 cases; communal females 56.1 ± 17.0 eggs, n “ 24 cases, t — 2.57, P < 0.05). In P. arizonica, no difference was observed in the number of eggs per egg case produced by solitary and communal females: solitary females, 21.5 ± 9.1 eggs per egg case, n = 69 cases; communal females, 22.2 ± 8.61 eggs, n — cases. Because females in this species make a large number of egg cases, it is difficult to be sure all egg cases are noted and collected; thus estimating the mean num- ber of egg cases per female is difficult. Given these caveats, there does not appear to be any significant difference in number of egg cases per female. For all females, the mean number of (observed) egg cases per female was 2.35 ±1.9 for solitary females {n = 40 females), 2.6 ± 2.7 for communal females {n = 25 fe- males; t = 0.44, P < 0.66, 63 df, two-tailed test, equal variances). For those females ob- served with at least one egg case, the figures are 2.9 ±1.6 egg cases per solitary female {n = 32 females), 3.61 ± 2.5 per communal fe- male (n = 18 females; t = 1.15, P < 0.26, 48 df). Colony structure: Both species occur in sol- itary webs and in aggregations. The aggrega- tions of both species contain two or more adult females, each with her own prey capture orb. The aggregations formed by the two spe- cies differ in several respects. Colonies of P. oweni attained larger size than those of P. arizonica: the largest P. ow~ eni colony observed contained 44 adult fe- males plus males and immatures. The mean number of females per web site at various lo- cations were: Cave Creek Canyon 1977, 3.5$ ± SD 8.6 (range 1-44, n = 25 web sites, 87$); South Fork Canyon, 1979, 1.7$ SD ± 2.0 (range 1-11, n = 52 web sites, 87$); and Herb Martyr, 1979, 1.3$ ± SD 0.6 (range 1- 4, n “ 40 web sites, 53$). In P. oweni colonies, orbs of adults and im- matures share support lines. The orbs are ar- ranged side by side in loose sheets of orbs, and several orbs or sheets of orbs may be stacked one over the other. Retreat(s) are not specially constructed by colony members; they are simply protected areas near the web such as a cleft under a rock or log surrounded by old webbing. The retreat or retreats may be used in common by all colony members. Aggregations of P. arizonica are smaller and simpler. The largest aggregation ever ob- served (in 1980) contained eight adult fe- males. In 1979 the mean number of females/ web site in South Fork Canyon was: 1.7 ± SD 1.2 (range 1-6, « = 45 web sites, 76$). Males and immatures may also be present in aggregations. The webs in an aggregation are side by side, joined by their space webbing. Retreats are usually absent. Feeding rates: Females of P. oweni spent a greater proportion of time feeding during the census period. On average 41.7 ± SD 17.8% of P. oweni females and 26.5 ± SD 11.7% of P. arizonica females were feeding per census hour (29 h, 10-13 P. oweni females, 17-20 P. arizonica females, P — 0.0009, Mann Whitney U-test). These measurements can be broken down to compare the feeding rates of solitary and aggregated females. In P. oweni an average of 53.1 ± SD 22.9% of communal females were feeding per census hour (6-8 fe- 16 THE JOURNAL OF ARACHNOLOGY Table 4. — Web measurements for Philoponella oweni and Philoponella arizonica; n = 22 webs of adult female Philoponella oweni, 19 webs of adult female P. arizonica. (Significance determined by two-tailed r-test for samples with equal variance). Parameter Mean SD P Longest diameter oweni 27.8 cm 10.6 <0.001 arizonica 12.4 3.8 Number of radii oweni 27.9 7.0 >0.05 arizonica 29.2 9.0 Number of spiral turns oweni 25.6 10.5 <0.005 arizonica 16.4 8.4 Deviation from horizontal oweni 51.9° 24.2 <0.001 arizonica 6.8° 13.8 Stabilimeetum present oweni 73% arizonica 0% males). This is significantly more than the time spent feeding by any other class of fe- males (29 h, P = 0.001 or less, Mann Whitney C/-test). There was no significant difference among the other three classes (P = 0.40 or more, Mann Whitney C/-test): solitary P. ow- eni, 22.9 ± SD 22.2% females feeding per hour, n = 2-5 females; aggregated P. arizon- ica, 28.7 ± SD 15.7% females feeding per hour, n = 7-10 females; solitary P. arizonica, 25.2 ± SD 15.4% feeding per hour, n = 7-11 females. Web structure: Web measurements are pre- sented in Table 4. Female P. oweni construct relatively large orbs which are closer to ver- tical than horizontal. There is a small quantity of space webbing below and around the orb, but there is seldom any above the orb. Stabi- limenta are usually present. The web P. ari- zonica consists of a small horizontal orb sur- rounded above, below and around the edges with space webbing. The orb is sometimes drawn up in the center by threads attached to the hub, giving it a slightly domed appear- ance. In many orbs the radials are not all in one plane, giving the orb a pleated appear- ance. None of these webs had stabilimenta. Habitats: In August 1980 all populations of P. oweni sampled were found at an elevation of 1630 m or higher (1630-1950 m) while all populations of P. arizonica were below 1630 m (1460-1630 m). The P. oweni webs were usually built in protected locations such as hollow trees and clefts between rocks. The P. arizonica webs were built in more open areas, such as in brush, shrubs or grass (Table 5). Egg-cases and immatures: The two species differ in the structure of their egg-cases. Phil- oponella oweni constructs large beige or co- coa-colored stellate egg-cases which are more-or-less flat on one side and domed on the other (Fig. 1). The mean length of the 24 egg cases measured was 6.7 mm (SD ± 0.96, range 5. 2-5. 8 mm); mean width was 4.4 mm (SD ± 0.61, range 2.9-5 .4 mm). These cases are heavily decorated with small spikes of Table 5. — Substrates used for web construction by Philoponella oweni and Philoponella arizonica: 1911 , Cave Creek population; 1979, South Fork Canyon populations. ® Yucca schottii absent from site. Rocks Yucca schottii Brush, shrubs Herbs, grass Base of trees Along logs Hollow trees Total 1977 oweni n 10 a 0 0 6 6 3 25 % 40 0 0 24 24 12 100 1979 oweni n 18 0 6 1 3 4 1 36 % 50 0 17 3 8 11 3 100 1979 arizonica n 3 15 6 6 1 0 0 31 % 10 48 19 19 3 0 0 100 SMITH— NOTES ON PHILOPONELLA 17 silk, especially on the curved side (spikes on flat side: median 5, range 0-22; spikes on curved side: median 13, range 5-24). Philo- ponella arizonica constructs pale smooth, whitish or bone colored egg^cases (Fig. 1). These cases are spindle-shaped, and there are usually no decorations or projections (of 33 egg-cases only eight were decorated with spikes, ranging in number from 7-12). Mean length of the 33 egg cases measured was 7.4 mm (SD ± 1.1, range 5. 1-9.1 mm); mean width was 2.8 mm (SD ± 0.30, range 2. 2-3.4 mm). The immatures of the two species are also recognizably different. P. oweni immatures are black with white markings and P. arizon- ica immatures are yellow with brown mark- ings. DISCUSSION Philoponella oweni and P. arizonica are found in close proximity in both time and space and occupy similar habitats. Although they are similar in appearance, they can easily be distinguished in the field by structure of the orb webs, nature of the communal groups (where they exist), the form of the egg cases and the coloration of second instar spiderlings. They also tend to use different substrates for web construction, with P. oweni making use of rigid substrates such as fallen logs, hollow trees, and niches under rocks, and P. arizonica making greater use of vegetation such as shrubs, grasses and yuccas as substrate. An enhanced food supply (whether due to higher prey capture rate, reduced prey han- dling time, increased size of prey, or other fac- tors) has often been proposed as a benefit of group-living behavior in spiders (e.g., B inford & Rypstra 1992, Buskirk 1975, 1981; Nen- twig 1985; Rypstra 1979, 1990). This study showed an interesting difference in feeding rates of the solitary and communal females P. oweni and P. arizonica. As was reported ear- lier (Snaith 1983), among P. oweni the pro- portion of females feeding per hour was great- er for communal than for solitary females. Insect trapping at the sites of communal and solitary P. oweni webs indicated that insect abundance was greater at sites occupied by colonies than at sites occupied by single webs. This suggests that communal groups are fea- sible at sites where insect abundance is high enough to support several females. Among P. arizonica, there was no differ- ence in the feeding rates of solitary and com- munal females — both were similar to the feed- ing rates of solitary P. oweni. No insect trapping was done in the vicinity of P. ari- zonica colonies and solitary webs, so it is not possible to say if insect abundance differs be- tween the sites of colonies and solitary webs. The earlier report on communal behavior of P. oweni also showed that females in com- munal groups produced a greater number of eggs per egg case than did solitary females, though total live young per female was the same for the two groups due to higher rates of egg case parasitism in the communal groups (Smith 1982). One explanation for this difference could be the difference in feeding rates between solitary and communal females. We observed no significant difference be- tween solitary and communal P. arizonica ei- ther in number of eggs per egg case or in egg cases per female, which dovetails with the feeding rates observed. However, as noted above, there are problems in collecting data on the number of egg cases per female in this species. Additional comparative study of the reproductive biology of communal and soli- tary Philoponella is warranted. Over their lifetimes, females of the two spe- cies produce the same average number of eggs and the same number of live second instar spi- derlings. However the two species differ in the way they package their eggs and care for the egg cases. P. oweni females package their eggs in one or a few large packets and make what appears to be a large expenditure in pa- rental care, in the form of guarding the egg- case without feeding. The females of P. ari- zonica, on the other hand, package their eggs into many small packets and continue to feed in their orbs while the egg-cases are suspend- ed in the web. Both species are subject to the same egg parasite, Arachnopteromalus dasys. It is not clear if the different egg-case tending behav- iors of P. oweni and P. arizonica have any effect against egg parasites such as Arachnop- teromalus dasys. One might suppose that the behavior of P. oweni affords more protection than that of P. arizonica. However another uloborid spider, Uloborus glomosus (Wal- ckenaer 1841), also makes several small-egg cases which it attaches to the web. In this spe- cies, the female has been observed to jerk the 18 THE JOURNAL OF ARACHNOLOGY web and make leg sweeping motions in re- sponse to parasitoid wasps (and spiderlings) crawling on the egg-cases (Cushing 1989; Cushing & Opell 1990), though it is not clear how effective this is in deterring parasites. I have observed A. dasys crawling on the egg- cases of both P. oweni and P. arizonica with no obvious reaction from the mothers of eggs. It is possible to make some testable hy- potheses concerning the adaptive significance (or lack of it) of the Philoponella egg-case tending behaviors. These hypotheses fall into four categories: those dealing with uncertain- ties faced by the female, those dealing with uncertainties faced by the young, those which consider differences in clutch size as side-ef- fects of other maternal behaviors, and non- adaptive explanations. Hypothesis I: P. arizonica females face more uncertainties in food supply than P. ow- eni females. When they gather enough re- sources for a small batch of eggs they produce a clutch right away; if they were to wait for additional prey they might use up their small reserve of energy in maintenance activities. This can be tested by measuring the feeding rates of marked individuals over time. P. ar- izonica females would be expected to have a higher variance in feeding rate than F. oweni females. Hypothesis 2: P. arizonica females axe sub- ject to a high and constant probability of mor- tality over their adult lives, while P. oweni females have relatively low probability dying before the first clutch is laid. It doesn't pay a P. arizonica female to save up resources for a large clutch if there is a good chance she will die before it can be laid. Life history data, particularly from the early part of the breeding season, are needed to test this hypothesis. Hypothesis 3: Females of P. oweni must guard their egg-cases because predators and egg-parasites are more common in their en- vironment than in that of P. arizonica. It would be more economical to produce a single large clutch than many small ones, since it takes as much time and energy to guard a small egg-case as a large one. This assumes that the type of maternal care shown by the bag species actually is more effective than that of the P. arizonica females in preventing par- asitism or predation. This can be tested by re- moving females from egg-cases, leaving the egg-cases in situ, and comparing the rates of parasitism on unguarded P. arizonica and P. oweni egg-cases to rates of parasitism on un- manipulated egg-cases. Hypothesis 4: P. arizonica is subject to a risky, unpredictable environment. The P. ari- zonica pattern of reproduction ensures that at least some of a female's offspring may hatch at a time when conditions are favorable. One obvious possibility is that spiderlings require a supply of very small prey, and that the avail- ability of these insects varies unpredictably over time. Little is known about the feeding behavior and survivorship of spiderlings. A first step would be to examine feeding behav- ior and prey of hatchlings, record variation in juvenile feeding rates over time, and correlate fluctuations in feeding rate with fluctuations in environmental factors such as rainfall. Hypothesis 5: The differences in reproduc- tive behaviors are not adaptations to any dif- ferences in ecology, behavior or microhabitat. Each species is conservative in behavior and displays the maternal behavior typical of its closest relatives. The first step in testing this hypothesis would be to construct a phytogeny for species in the Philoponella semiplumosa species group (Opell 1979, 1987), and exam- ine the maternal behavior of the closest rela- tives of P. oweni and P. arizonica species. ACKNOWLEDGMENTS I thank Brent Opell for identification of Philoponella oweni and P. arizonica speci- mens, and Willis Gertsch and Vincent Roth for encouragement and advice while I was working at the Southwestern Research station. Robert Hagen provided useful criticism of the manuscript. This research was supported in part by grants from the Theodore Roosevelt Memorial Fund of the American Museum, Sigma Xi National, and Sigma Xi Cornell chapter. Long ago, this study was part of my doctoral research at Cornell University. I am sorry that my advisor and friend, Dr. George C. Eickwort, did not live to see the last chap- ter of my thesis published. LITERATURE CITED Binford, G. & A.L. Rypstra. 1992. Foraging be- havior of the communal spider, Philoponella re- publicana (Araneae: Uloboridae). J. Insect Be- hav., 5:321-355. Breitwisch, R. 1989. Prey capture by a West Af- rican social spider (Uloboridae: Philoponella sp.). Biotropica, 21:359-363. SMITH— NOTES ON PHILOPONELLA 19 Buskirk, R.E. 1975. Coloniality, activity patterns and feeding in a tropical orb-weaving spider. Ecology, 56:1314-1328. Buskirk, R.E. 1981. Sociality in the Arachnida, Pp. 282-367, In Social Insects, Vol. II.(H.R. Herman, ed.). Academic Press, New York. Cushing, PE. 1989. Possible egg sac defense be- haviors in the spider Uloborus glomosus (Ara- neae: Uloboridae). Psyche, 96:269-211 . Cushing, RE. & B.D. Opell. 1990. Disturbance be- haviors in the spider Uloborus glomosus (Ara- neae, Uloboridae): possible predator avoidance strategies. Canadian J. ZooL, 68:1090-1097. Eberhard, W.G. 1977. The webs of newly emerged Uloborus diversus and of a male Uloborus sp. (Araneae: Uloboridae). J. Arachnol., 4:201-206. Gordh, G. 1976. A new genus of Pteromalidae from Missouri, the type species of which para- sitizes Uloborus octonarius Muma (Hymenop- tera: Chalcidoidea; Araneida: Uloboridae). J, Kansas Ent. Soc., 49:100-104. Lahmann, E.J. & W.G. Eberhard. 1979. Factores selectivos que afectan la tendencia a agruparse en la arana colonial Philoponella semiplumosa (Araneae; Uloboridae). Rev. BioL Trop., 27:231- 240. Muma, M.M. & W.J. Gertsch. 1964. The spider family Uloboridae in North America north of Mexico. Amer. Mus. Novitates, 2196:1-43. Nentwig, W. 1985. Social spiders catch larger prey: a study of Anelosimus eximius (Araneae: Theri- diidae). Behav. Ecol. SociobioL, 17:79-85. Opell, B.D. 1979. Revision of the genera and trop- ical American species of the spider family Ulo- boridae. Bull. Mus. Comp. ZooL, 148:443-549. Opell, B.D. 1987. The new species Philoponella herediae and its modified orb web (Araneae, Uloboridae). J. Arachnol., 15:59-63. Rypstra, A.L. 1979. Foraging flocks of spiders: A study of aggregate behavior in Cyrtophora citri- cola Forskal (Araneae; Araneidae) in West Af- rica. Behav. Ecol. SociobioL, 5:291-300. Rypstra, A.L. 1990. Prey capture and feeding ef- ficiency of social and solitary spiders: a compar- ison. Acta ZooL Fennica, 190:339-343. Smith, D.R. 1982. Reproductive success of solitary and communal Philoponella oweni (Araneae: Uloboridae). Behav. Ecol. SociobioL, 11:149- 154. Smith, D.R. 1983. Ecological costs and benefits of communal behavior in a presocial spider. Behav. Ecol. SociobioL, 13:107-114. Smith, D.R. 1985. Habitat use by colonies of Phil- oponella republicana (Araneae: Uloboridae). J. Arachnol., 13:363-373. Szlep, R. 1961. Developmental changes in the web-spinning instinct of Uloboridae: construc- tion of the primary type web. Behaviour, 27:60- 70. Received 28 February 1995, revised 10 October 1996. 1997. The Journal of Arachnology 25:20-30 BEHAVIOR AND NICHE SELECTION BY MAILBOX SPIDERS Robert L. Edwards^ and Eric H* Edwards^: ^Box 505, Woods Hole, Massachusetts 02543 USA and ^45 Canterbury Lane, East Falmouth, Massachusetts 02536 USA ABSTRACT. The data for species of spiders observed and collected for a period of eight years from a rural delivery mailbox route in Mashpee, Massachusetts is examined. We collected 1252 individuals, with 199 species represented. Some species were year-round residents of mailboxes while others appeared only during limited periods of time. Species typically found in the foliage of coniferous trees and on the tranks of pines and oaks dominated the collections, with lesser numbers from other types of habitats. The species observed are divided into categories depending upon their consistency in terms of time of occurrence and number. Species that occurred only rarely tended to be different from year to year. Arachnologists have long been aware that the structure of the habitat, along with sea^ sonal and other environmental factors, plays a dominant role in determining where spiders are to be found (Stratton, Uetz & Dillery 1978; Hatley & MacMahon 1980; Bultman & Uetz 1983; Gunnarsson 1983, 1992; Green- stone 1984; Rypstra 1986; Moring & Stewart 1994; Reichert & Gillespie 1986; Rushton 1991; Sundberg & Gunnarsson 1994). Defining the niche of any organism is a daunting task. Each species has a complex set of interacting biotic and abiotic requirements within which it survives (Hutchinson 1957). In the case of spiders it is difficult to define their individual niche requirements based only on the specific habitat within which they have been collected. A surprising number of spe- cies collected in well defined habitats are clearly not typical occupants of the habitat and may be considered rare or accidental. When the sampling procedure is based on a set of quadrats, a species that occurs in only one quadrat, whatever the number of individuals, is referred to as a 'unique' species {cf. Helts- che & Forrester 1984). The term 'unique' is neutral in that it does not presume that the species is necessarily rare or accidental within the habitat. Unique species make up a large percentage of the spider species collected in many habitats, varying from 25-50% of the total number of species collected (Edwards 1993). Spiders are vagile and accordingly tend to confuse the issue when one is attempting to describe a typical species assemblage for any particular habitat. Some insight may be gained into the nature of unique species, the niche-spatial requirements of spider species and by the species assemblages observed from an examination of the data obtained collecting spiders from an artificial habitat; in this case the rural delivery mailboxes in Mashpee, Barnstable County, Massachusetts. METHODS Typical mailboxes and their settings are shown in Figs. 1, 2. The standard box is made of galvanized sheet metal, usually 16.5 cm wide, 21.5 cm high and 48 cm long and has a rounded top (Fig. 1). The mailbox is often painted black or variously decorated by the owner. The box is supported by a pipe or stout post, circa 8 cm in cross section upon which it is directly seated or from which it is canti- levered and may have additional oblique sup- ports at the bottom (Fig. 2). On sunny days these boxes may get very warm. Attendance to 350-400 such boxes, involving some 40 km of travel daily, Monday-Saturday, com- prises the average route. The mailman, Eric Edwards, is familiar with the local species and collects those spiders not previously collected, or that had not been collected in any particular month. Time constraints and other factors make it impossible to observe or collect spi- ders from these boxes systematically. The mailbox is described and the results of the ini- tial three years of data collection are provided in Edwards & Edwards (1991). As of July 1995, eight years of collecting and observa- tion have been completed and 199 species (1252 individuals) of spiders collected. The 20 EDWARDS & EDWARDS— MAILBOX SPIDERS 21 Figures 1, 2. — Photographs of rural delivery mailboxes. 1, Box fastened to top of post, front end with door open. The projecting handle at top of door and handle lock on top of box. Note space between bottom of door and box; 2, Cantilevered mailbox. Notice that post projects above the mailbox and the oblique support beneath. mailboxes are usually situated a short distance away from vegetation other than short grass or lawn. Occasionally there will be a simple, doorless box on a slender metal stake nearby for newspaper deliveries. These boxes were not sampled. The area has many ponds and bogs, some fields, and abundant lawns, with pitch pine (Pinus rigida Mich.), white pine (Pinus strobus L.), red cedar {Juniperus vir~ ginianus L.) and several species of oaks (red -Quercus rubra L.; scarlet -Q. coccinea Muenshh.; and white ~Q. alba L.) dominating the patches of woods in the surrounding areas. A large variety of shrubs, both local species and horticultural varieties, are found nearby. The mailboxes offer a unique set of spatial options to the spiders that happen upon them. These options include the outer, smooth sur- face, approximately 3,670 cm^, the dark inte- rior of the box, the handle and door lock that extend up and out from the box when the door is closed, the outer bottom surface, and any space between the overlapping flange of the door and the box itself on the sides and top. The space between the bottom of the door and the box is fairly wide (±5 mm), and is used as passages by many species (Fig. 1). Other than spiders, prey in the form of ants and flies are frequently found on the box. Representa- tive collections of species have been deposited in the United States National Museum. As in the case of agroecosystems (Rypstra & Carter 1995), the mailboxes are newly col- onized each year with a large number of spe- cies that have overwintered elsewhere. The niche-spatial options offered by the mailbox represent a consistent set of microhabitats within an artificial habitat that, in turn, exists within a complex array of natural habitats. RESULTS AND DISCUSSION Unique species. — ^Considering each month as a separate quadrat for the purposes of this study, 72 species (36%) of spiders collected from the mailboxes during the period June 1987-July 1995 classify as unique species (Table 1). Sixty-five of these were represented by single individuals, seven by two individu- als. The two seasonal modes in the number of species, early summer and fall, are typical of the overall area. The unique species are roughly proportional to the total number of 22 THE JOURNAL OF ARACHNOLOGY Table L — Number of species collected in one month only during the period June 1987-July 1995. Collected from Mashpee, Massachusetts rural delivery mailboxes. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 1987 4 2 1 1 8 1988 2 1 3 1989 2 3 6 5 1 3 2 3 25 1990 1 1 2 1 5 1991 2 1 3 1 2 1 10 1992 1 1 4 1 7 1993 1 3 1 5 1994 1 3 1 1 1 7 1995 1 1 2 Total 1 0 3 5 8 20 13 5 7 6 3 1 72 species found each month (Fig. 3), suggesting that ballooning accounts for many of these oc= currences (Bishop & Riechert 1990). Over a long period of time one would expect the number of these species to diminish slowly as, by virtue of their vagility, individuals of all the species of the regional pool will eventually happen upon the mailbox. The regional pool of spiders in the Mashpee area is estimated to be approximately 500 species (Edwards 1993). Residential species. — At least 39 species are found on the mailboxes much of the year and are referred to here as ’residential’ species (Table 2). The boxes are disturbed to some degree on most mail days because of the large amounts of material gratuitously sent to ’Box Holder’ or ’Resident Box RR04.’ From time to time, large numbers of boxes are vandal- ized. In spite of this, many species establish more or less permanent positions with capture Figure 3. — The total number of species and num- ber of unique species collected from Mashpee mail- boxes, by month from June 1987=July 1995. webs and/or retreats in or on the box (Ed- wards & Edwards 1991). For example, Stea- toda borealis (Hentz 1850) is consistently found deep inside the box where it builds its web and deposits its egg sacs. In a natural setting this spider is found in recesses in the trunks of trees and logs, but can be common also under domestic refuse, such as piles of old lumber around houses. Similarly, Pityoh- yphantes costatus (Hentz 1850) maintains a sheet web near the front end at the top of the box where it is also less affected by the act of delivering or removing mail. Other species, for example Philodromus vulgaris (Hentz 1847), build retreats (within which egg sacs are deposited) along the inside edges of the door and box and search for prey outside. These species move in and out of the box free- ly through the space at the bottom and around any other open edges of the door. Three ther- idiid species, Thymoites unimaculatus (Emer- ton 1882), Theridion murarium Emerton 1882, and Theridion lyricum Walckenaer 1841, and the tetragnathid Tetragnatha viridis Walckenaer 1841, are consistently found on the upward and outward projecting handle and door lock of the box (Fig. 1). The theridiids maintain webbing here, apparently replacing it readily despite disturbance. T. viridis makes no obvious organized web and seems to be- have more like a mimetid spider: its presence appears to discourage the close presence of other species. We have found T. viridis on pines and cedars both day and night, and in all cases also without obvious capture webs (orbs). Uloborus glomosus (Walckenaer 1841) was observed only on white boxes, and was one of only a very few species that construct- EDWARDS & EDWARDS—MAILBOX SPIDERS 23 Table 2. — Mailbox species considered as residential. Based on collections and observations, June 1987- July 1995, Mashpee, Massachusetts. Arranged on basis of life cycle stages represented. Juv. = juvenile, Ad. = adult. The natural habitat in the Mashpee area for these species is indicated. Species Months Stage Habitat Anyphaena ceier (Hentz) Mar-Oct Juv. & Ad. conifer Hibana gracilis (Hentz) Apr-Nov Juv. & Ad. conifer Araneus bivittatus (Walckenaer) Jan-Dec Juv. & Ad. conifer Araneus gadus Levi Mar-Dec Juv. & Ad. conifer Araniella displicata (Hentz) Apr-Dee Juv. & Ad. conifer Eustala anastera (Walckenaer) Jan-Dec Juv. & Ad. conifer Grammonota pictilis (O.P. — Cambridge) Mar-No V Juv, & Ad. conifer Pityohyphantes costatus (Hentz) Jan-Dec Juv. & Ad. conifer Mimetus notius Chamberlin Feb-Nov Juv. & Ad. conifer Oxyopes scalaris Hentz Jan-Dec Juv. & Ad. conifer Philodromus rufus Walckenaer Mar-Dec Juv. & Ad. conifer Metaphidippus exiguus (Banks) May-Oct Juv. & Ad. conifer Xysticus punctatus Keyserling Max-Dec Juv. & Ad. conifer Theridion lyricum Walckenaer Jan-Dec Juv. & Ad. conifer Theridion murarium Emerton Jan-Dec Juv. & Ad. conifer Thymoites unimaculatus (Emerton) May-No V Juv. & Ad. conifer Clubionoides excepta (C.L. Koch) May-Sep Juv. & Ad. trunk Philodromus vulgaris (Hentz) Feb-Dec Juv. & Ad. trunk Maevia vittata (Hentz) Apr-Oct Juv. & Ad. trunk Rhidippus audax (Hentz) Mar-No v Juv. & Ad. trunk Platycryptus undata (DeGeer) May-Sep Juv. & Ad. trunk Coriarachne versicolor Keyserling Mar-Sep Juv. & Ad. trunk Steatoda borealis (Hentz) Apr-Dee Juv. & Ad. trunk Theridion lyricum Walckenaer Jan-Dec Juv. & Ad. trank Euryopis limbata (Walckenaer) Jun-Sep Juv. & Ad. trank Leucauge venusta (Walckenaer) Apr-Nov Juv. & Ad. understory Philodromus marxi Keyserling Apr-Sep Juv. & Ad. understory Tmarus angulatus (Walckenaer) Apr-Nov Juv. & Ad. understory Uloborus glomosus (Walckenaer) May-Nov Juv. & Ad. understory Metaphidippus protervus (Walckenaer) Mar-Dec Juv. & Ad. field Misumenops asperatus (Hentz) Apr-Nov Juv. & Ad. field Achaearanea tepidariorum (C.L. Koch) Jan-Dec Juv. & Ad. house Ceratinopsis atolma Emerton Jan-Nov Adults only conifer Ceratinopsis nigripalpis Emerton Jan-Dec Adults only understory Ceratinops lata (Emerton) Apr-Sep Adults only trank Eperigone maculata (Banks) Feb-Nov Adults only litter Philodromus laticeps Keyserling Mar-No v Juv. only conifer Tetragnatha versicolor Walckenaer Jan-Dec Juv. only field Tetragnatha viridis Walckenaer Jan-Nov Juv. only conifer ed well-defined orb webs on the box. Phidip- pus audax (Hentz 1845) is frequently encoun- tered on the outside of the box and in retreats inside with egg sacs. One box had this species present throughout the sampling period. Although egg sacs were frequently found, most could not be positively identified to spe- cies. However, it is clear that not all the spe- cies categorized as residential fully completed their life cycle on the box. Three species pres- ent much of the year, Philodromus laticeps Keyserling 1880, Tetragnatha viridis and T. versicolor Walckenaer 1841, leave the mail- boxes as adults, presumably to mate and de- posit egg sacs elsewhere. Both juveniles and adults of Coriarachne versicolor Keyserling 1880, (a darkly-colored crab spider, usually taken on pine tree trunks) and Xysticus punc- tatus Keyserling 1880, (a lightly-colored crab spider found in the foliage of conifers) are found on the boxes. These two species are found in the open in their natural habitat dur- 24 THE JOURNAL OF ARACHNOLOGY ing the day. Three erigonine species, Ceratin- ops lata (Emerton 1882), Ceratinopsis atolma Chamberlin 1925, and Ceratinopsis nigripai- pis Emerton 1882, are present only as adults. The residential category as a whole is domi- nated by species most likely to be taken in coniferous foliage and on tree tranks (Table 2). Seasonal species. ^—Thirty-five species, here categorized as 'seasoeaF species, oc- curred consistently on the mailboxes for pe- riods of 2-4 months, or occasionally more, during the year (Table 3). This group is dom- inated by species represented mostly, if not entirely, by adults. Species normally taken in the forest understory dominate. Some are warm weather species, others cold weather species. The population of Micrathena sagi- tatta (Walckenaer 1841) dramatically in- creased in recent years (1994-1995). Adults and a few late instars were encountered more frequently at this time, on the boxes and in webs anchored between the top edge of the box and the upper end of the supporting post (Fig. 2) or between the handle and the lower surface of the door. Xysticus fraternus Banks 1895, is most often found in leaf litter but shows up on boxes only as adults in June and Table 3. — Species found seasonally on mailboxes. Arrayed by life cycle stages. Ad. = adult, Juv. = juvenile, Adults + , males + , and females + indicates very few juveniles also found. Juv.+ indicates very few adults collected or observed. The natural habitat for these species in the Nashpee, Massachusetts area is indicated. Species Months Stage Habitat Eris militaris (Hentz) Apr-Jun Juv. & Ad. understory Hentzia mitrata (Hentz) May-Jun Juv. & Ad. understory Admestina wheeleri (Peck. & Peck.) May-Jun Juv. & Ad. trunk Herpyllus ecciesiasticus Hentz Jul-Oct Juv. & Ad. trunk Metaphidippus insignis (Banks) May-Jul Juv. & Ad. field Saiticus scenicus (L.) Apr-Jul Juv. & Ad. field Tutelina similis (Banks) Jun-Sep Juv. & Ad. field Anyphaena pectorosa L. Koch Jun-Aug Adults T understory Micrathena sagittata (Walckenaer) Jun-Sep AdultsE understory Gladicosa pulchra (Keyserling) Aug-Oct Adults + trunk Hyposinga rubens (Hentz) Jun-Jul Adults T trunk Neoscona arabesca (Walckenaer) Jul-Aug AdultsT field Centromerus latidens (Emerton) Mar-May Adults + litter Ceraticelus alticeps (Fox) May-Jul Ad. only conifer Theridion glaucescens Becker Jun-Aug Ad. only conifer Agelenopsis potteri (Blackwall) Aug-Sep Ad. only understory Argyrodes trigonum (Hentz) Jul-Sep Ad. only understory Origanates rostratus (Emerton) Dec-Apr Ad. only understory Sciastes truncatus (Emerton) Oct-Mar Ad. only understory Souigas corticarius (Emerton) Sep-Jan Ad. only trank Erigone autumnalis Emerton Jan- Aug Ad. only field Erigone dentigera (O.P.— Cambridge) Jun-Jul Ad. only field Xysticus fraternus Banks Jun-Jul Ad. only litter Robertus pumilus (Emerton) Mar-Apr Females + litter Agelenopsis pennsyivanicus (C.L.K.) Aug-Oct Females understory Trachelus tranquillus (Hentz) Oct-Nov Females mailbox Dictyna minuta Emerton May-Jul Males conifer Frontinella pyramitela (Walckenaer) Mar-No V Males conifer Zyballus bettini Peckham Jan-Sep Males field Meriene clathrata (Sundevall) Apr Juv.+ litter Steatoda americana (Emerton) Apr-Sep Juv.+ litter Dipoena nigra (Emerton) Jun-Jul Juv. trank Hentzia palmarum (Hentz) Apr-Oct Juv. understory Pisaurina mira (Walckenaer) Aug-Oct Juv. understory Trabeops aurantiaca (Emerton) Mar-Jun Juv. litter EDWARDS & EDWARDS— MAILBOX SPIDERS July. Two species of Agelenopsis, pennsylvan- icus (C.L. Koch 1843) and potteri (Blackwall 1846), appear briefly in late summer and early fall as adult females and deposit egg sacs in the mailbox. This is consistent with their be- havior in natural settings. As they mature they tend to build larger and higher funnel webs in the understory, and frequently deposit their egg sacs under loose bark or other such re- fugia. In one unusual circumstance, in a web shared by both a male and female A. penn- sylvanicus, a female Trachelus tranquillus (Hentz 1847), had been captured. Adult fe- males of Trachelus tranquillus consistently show up in the mailbox only in the fall. Tra- beops aurantiaca (Emerton 1885) took refuge inside the box in late spring as preadult in- stars. At this time of the year they are taken high up on understory shrubbery, possibly as a prelude to ballooning: otherwise they tend to be found most commonly on the forest floor and in leaf litter. Several erigonine species of the genera Erigone, Eperigone, Grammonota and Walckenaeria can be abundant in lawns. Of these only adults of Erigone autumnalis (Emerton 1882) and Erigone dentigera (O.P.- Cambridge 1874) showed up regularly on the boxes. The adults of two other erigonine spe- cies, Ceratinops lata (Emerton 1882), and Soulgas corticarius (Emerton 1909) are found in the narrow space between the overlapping rim of the door and the box. They take shelter under shallow, shaded refuges on tree trunks such as those provided by lichens. Six of the seven seasonal species that included a range of instars (juveniles and adults) are salticids. Ballooning. — The large number of random strands of silk observed at the uppermost part of the box, the handle, and the top of the post suggest that these positions were used as launching points for ballooning. However, no spiders were observed in the act of ballooning. This activity may account for the presence of some species, particularly those in the unique and unassigned categories. Distribution trends.— There are clear trends in the numbers of species from the unique to residential categories (Table 4). Of the 199 species collected on the mailbox, 125 are represented by few records and/or sporadic occurrence and could not be assigned to either the residential or seasonal categories with any confidence (72 unique, 53 unassigned). The unique category was dominated by species 25 Table 4. — Habitats where species collected from mailboxes are most likely to be found in the Mash- pee, Massachusetts area, for each category used in text. Unique Unas- signed Seasonal Residen- tial Field 36 19 7 3 Leaf litter 13 10 6 1 Understory 5 15 11 5 Conifer 5 5 4 19 Tree trank 3 1 6 10 Around house 3 2 1 Mailbox only 7 1 1 Totals 72 53 35 39 commonly found in fields and leaf litter. Spe- cies from such habitats decreased in number sequentially to just a few in the residential cat- egory. Species only taken on mailboxes are particularly interesting since, so far, they still remain to be taken elsewhere in this area de- spite intensive collecting over many years. Some examples include Ceraticelus bryantae Kaston 1945, reported from Connecticut; Marpissa wallacei Barnes 1958, which has yet to be reported further north than Georgia, and Disembolus sacerdotalis Crosby & Bishop 1933, apparently a rare species known only from the holotype (Millidge 1981). Few un- derstory and coniferous species occurred as unique species with the exception of the larger species of Araneus. Araneus probably found little support in the immediate vicinity of the box for constructing orbs. Eustala anastera (Walckenaer 1841), on the other hand, is found year round on the box, but without web- bing. Comments on source habitats.— -The un- assigned category is dominated by species usually found in fields and on understory fo- liage (Table 4). Understory species dominated the seasonal category. The residential cate- gory is made up largely of species (74%) typ- ically found on two types of natural habitats, coniferous foliage and tree trunks. Thirty-nine (53%) of the 74 species in the seasonal and residental categories are taken on coniferous foliage and tree trunks. These last two habitat types are the principal sources of the consis- tently observed mailbox spiders. Table 5 lists 104 species taken in coniferous foliage (pitch pine and red cedar) and on the 26 THE JOURNAL OF ARACHNOLOGY Table 5. — Percent of quadrats occupied by species in foliage of pitch pine and red cedar and on trunks of pitch pine and oaks. Study carried out on Cape Cod, Massachusetts, 1989 and 1990 (Edwards 1993). Arrayed as categorized for species taken from mailboxes, and within each category in order by those taken on coniferous foliage only, on both foliage and trunks, and on trunks only. Pine Foliage Cedar Pine Trunk Oak No. of species 67 65 42 44 No. of quadrats 40 44 40 41 No. of unique species 23 15 19 22 Species not taken on mailbox Cesonia bilineata (Hentz) 2.3 Ceraticelus pygmaeus (Emerton) 2.5 Misumenops formosipes (Walckenaer) 2.5 Thanatus formicinus (Olivier) 2.5 Walckenaeria brevicornis (Emerton) 2.5 Philodromus pernix Blackwall 4.5 Neriene radiata (Walckenaer) 4.5 Phoroncidia americana (Emerton) 2.5 2.3 Steatoda albomaculata (DeGeer) 2.5 2.3 Grammonota maculata Banks 5.0 Grammonota ornata (O.P. — Cambridge) 5.0 Litopyllus temporarius Chamberlin 2.5 2.5 2.4 Achaearanea globosum (Hentz) 10.0 34.1 2.4 Sergiolus variegatus (Hentz) 2.4 Araneus pratensis (Emerton) 2.5 Species categorized as unique on mailbox Araneus diadematus Clerck 2.5 Dipoena buccalis Keyserling 2.5 Araneus miniatus (Walckenaer) 2.5 15.9 2.5 Theridion frondeum Hentz 2.5 4.6 2.5 12.2 Theridion crispulum Simon 12.5 25.0 2.5 4.9 Mangora gibberosa (Hentz) 2.4 Strotarchus picatorius (Hentz) 31.7 Drapetisca alteranda Chamberlin 5.0 70.7 Species categorized as unassigned on mailbox Ceraticelus similus (Banks) 2.3 Ero leonina (Hentz) 2.3 Philodromus placidus Banks 2.3 Mangora placida (Hentz) 2.5 2.3 Phidippus whitmani (Peckham) 2.5 2.3 Gonatium crassipalpum Bryant 6.8 Philodromus exilis Banks 2.5 6.8 Theridion dijferens Emerton 5.0 2.3 Hypselistes florens (O.P. — Cambridge) 5.0 4.5 Cyclosa conica (Pallas) 10.0 Araneus marmoreus Clerck 11.4 Mangora maculata (Keyserling) 11.4 Enoplognatha ovata (Clerck) 5.0 6.8 Philodromus imbecillus Keyserling 12.5 2.3 Eris pineus (Kaston) 22.5 Araneus partitus (Walckenaer) 2.5 2.3 2.4 Theridion alabamense Gertsch & Archer 5.0 13.6 25.0 36.6 Nodocion floridanus (Banks) 2.4 Sylaceus pallidus (Emerton) 2.4 Tetragnatha laboriosa Hentz 2.4 EDWARDS & EDWARDS— MAILBOX SPIDERS 27 Table 5. — Continued. Foliage Trunk Pine Cedar Pine Oak Tutelina elegans (Hentz) 2.5 Mimetus puritanus Chamberlin 2.5 Lepthyphantes sabulosa (Keyserling) 2.5 Tutelina similus (Banks) 5.0 Pulex habrocestum (Hentz) 2.5 17.1 Philodromus validus (Gertsch) 60.0 2.4 Species categorized as seasonal on mailbox Dictyna minuta Emerton 5.0 Neoscona arabesca (Walckenaer) 5.0 Theridion glaucescens Becker 7.5 Eris militaris (Hentz) 5.0 9.1 Pisaurina mira (Walckenaer) 12.5 4.5 Zygoballus bettini Peckham 2.3 2.4 Centromerus latidens (Emerton) 4.5 2.5 2.4 Anyphaena pectorosa L. Koch 5.0 13.6 2.5 2.4 Argyrodes trigonum (Hentz) 20.0 13.6 7.3 Gladicosa pulchra (Keyserling) 2.5 15.0 22.0 Frontinella pyramitela (Walckenaer) 37.5 36.4 7.5 2.4 Hypsosinga rubens (Hentz) 2.3 57.5 39.0 Dipoena nigra (Emerton) 30.0 9.1 45.0 39.0 Ceraticelus alticeps (Fox) 82.5 59.1 45.0 9.8 Agelenopsis pennsylvanicus (C.L. Koch) 2.4 Hentzia mitrata (Hentz) 2.4 Hentzia palmarum (Hentz) 2.4 Admestina wheeleri Peckham & Peckham 15.0 12.2 Herpyllus ecclesiasticus Hentz Species categorized as residential on mailbox 27.5 7.3 Ceratinopsis nigripalpis Emerton 2.3 Eperigone maculata (Banks) 2.3 Tetragnatha versicolor Walckenaer Maevia vittata (Hentz) 2.5 2.3 Ceratinopsis atolma Chamberlin 2.5 4.5 Philodromus laticeps Keyserling 17.5 Araniella displicata (Hentz) 2.5 15.9 Oxyopes scalaris Hentz 20.0 15.9 Tetragnatha viridis Walckenaer 15.0 29.5 Philodromus rufus Walckenaer 56.8 Mimetus notius Chamberlin 25.0 22.7 Anyphaena celer (Hentz) 25.0 47.7 Xysticus punctatus Keyserling 55.0 40.9 Misumenops asperatus (Hentz) 2.5 4.5 2.5 Tmarus angulatus (Walckenaer) 2.5 6.8 5.0 Uloborus glomosus (Walckenaer) 10.0 4.5 2.5 2.4 Pityohyphantes costatus (Hentz) 5.0 13.6 2.5 Araneus gadus Levi 12.5 11.4 2.4 Soulgas corticarius (Emerton) 2.5 17.5 9.8 Philodromus marxi Keyserling 10.0 18.2 2.4 Metaphidippus protervus (Walckenaer) 10.0 18.2 2.5 Euryopis limbata (Walckenaer) 5.0 4.5 5.0 22.0 Leucauge venusta (Walckenaer) 12.5 11.4 2.5 12.2 Hibana gracilis (Hentz) 40.0 4.5 2.5 Philodromus vulgaris (Hentz) 22.5 18.2 5.0 4.9 Coriarachne versicolor Keyserling 2.5 2.3 50.0 4,9 28 THE JOURNAL OF ARACHNOLOGY Table 5. — Continued. Foliage Trunk Pine Cedar Pine Oak Thymoites unimaculatum (Emerton) Araneus bivittatus (Walckenaer) Eustala anastera (Walckenaer) Metaphidippus exiguus (Banks) Theridion murarium Emerton Grammonota pictilis (O.R — Cambridge) Clubionoides excepta (L. Koch) Theridion lyricum Walckenaer Playcryptus undata (DeGeer) Ceratinops lata (Emerton) 17.5 25.0 20.0 2.4 32.5 38.6 5.0 27.5 34.1 10.0 22.0 50.0 50.0 2.4 72.5 Al.l 2.5 2.4 52.5 81.8 2.5 25.0 27.3 55.0 43.9 32.5 56.8 50.0 75.6 15.0 15.0 34.1 trunks of pitch pine and of oaks (red, scarlet and white). For each quadrat in these habitats, I was sampled by beating (coniferous fo- liage); brushing (oak trunks) and bark removal (pitch pine). For further information on col- lection methods used and a description of these habitats, see Edwards (1993). Of the 104 species taken, 15 (14%) were not collected on the mailboxes; and, of these, I I occurred in coniferous foliage only (Table 6). With the exception of Achaearanea glo- bosum (Hentz 1850), these species were rep- resented only in a small percentage of quad- rats, suggesting that they occurred accidentally or were uncommon. Only eight (11%) of the 72 unique species found on the mailboxes occurred in either of the principal source habitats, with no other particular out- side source suggested (Table 6). Categorized as a unique species on the mailbox, Strotar- chus piscatorius (Hentz 1847) was taken only on the trunks of oaks where adult females Table 6. — Distribution of mailbox species taken from tranks only, from coniferous foliage only, and from both foliage and trunks. T = trunk, F = fo- liage, TF = both trank and foliage, n = total num- ber of species. Mailbox category T F TF n Residential 2 13 21 36 Seasonal 5 4 10 19 Unassigned 9 15 2 26 Unique 3 2 3 8 Not on mailbox 2 11 2 15 Total 20 45 39 104 with egg sacs were found in shaded, moist crevices. A comparable niche option was not offered by the mailbox. The majority of the 26 species in the un- assigned category were also not abundant in any of the four natural habitats. Fifteen spe- cies were found in foliage habitats only, nine solely from trunk habitats, and just two on both types of habitats (Table 6), suggesting that some species with more restricted niches tend not to be attracted to the mailbox. Eight unassigned species (31%) were found in more than 10% of the quadrats in natural habitats (Table 5), although most were confined to ei- ther foliage or trunks with the exception of Theridion alabamense Gertsch & Archer 1942 (25.0% of pine trunks and 36.6% of oak trunks). Particularly interesting are two spe- cies that are taken abundantly and only on the trunks of pine and smooth barked trues such as oak and beech. Drapetisca alteranda Chamberlin 1909 and Philodromus validus (Gertsch 1993) have been taken on mailboxes once and twice respectively. D. alteranda is one of the most abundant species collected from the relatively smooth barked oak trees (70.7% of quadrats). It produces a flimsy sheet web, vaguely circular in outline. The webbing tends to be supported by minor projections of the bark, otherwise it is essentially flat. The spider sits anchored in a depression, usually at the periphery of the web. As a consequence of its being anchored, when using a stiff brush as a sampling tool one often collects only the cephalothorax. It is unclear how this spider fixes itself to the bark. The mailbox did not offer a comparable setting. Philodromus vaT EDWARDS & EDWARDS— MAILBOX SPIDERS idus was the most common spider taken on pitch pine trunks, 60.0% of quadrats. It ap- pears to prefer the rough-barked pitch pine where it takes refuge during the day in the many shallow leaf-like crevices of the bark. Again, the mailbox did not provide a com- parable spatial niche. Here, again, it appears that specialization, e.g., trunks as opposed to foliage, tends to limit occurrence on the mail- box. Of the 35 species in the mailbox seasonal category, 19 (68%) were taken in the four principal source habitats, ten of which were taken both in coniferous foliage and on pine and oak trunks (Table 6). It should be noted that the seasonal category (Table 5) includes many more abundant species than those in- cluded in the previous categories. However, one notable exception is the more abundant spider on coniferous foliage and pitch pine trunks, Ceraticelus alticeps (Fox 1891), found in 56.0-82.5% of quadrats. This small erigon- ine species occurs as well in the foliage of deciduous trees. It has been taken on mail- boxes as adults only, not abundantly, suggest- ing a preference for truly arboreal situations. In contrast, juveniles and adults of a slightly larger erigonine species, Grammonota pictilis (O.P. -Cambridge 1875) also abundant in co- niferous foliage, are to be found on the mail- box much of the year and categorized as res- idential. Agelenopsis pennsylvanicus and A. potteri females, as noted earlier, consistently appear inside the mailboxes in late summer, where they construct sheet webs both in and out of the box, and deposit their egg sacs. The collecting method used (brushing) on oak trunks is not an effective method for collect- ing these two species or any other spider that tends to hide underneath large pieces of dead bark. Aside from the spiders that built webs on the handle and the salticids, many of the erigonine species were found in retreats in spaces between the door flange and sides of the box or just inside the box where the floor meets the wall. Eight species (42%) were present in 10% or more of the quadrats of the natural habitats. Most of the species in the residential cate- gory are represented in these natural habitats by relatively abundant spiders. Twenty-one were to be found in both foliage and trunk habitats, and 13 in coniferous foliage only, suggesting that the former group were more 29 eclectic in selecting a “home” or prone to moving about. Twenty-four (83%) of the 36 residential species were taken in more than 10% of the quadrats in one or more of the natural habitats. It will be noted that 72 unique species were collected from the mailboxes (36% of total), and approximately the same percentage, 34% (23 unique species), from pitch pine foliage; and somewhat less, 23% (15 unique species), from red cedar foliage. On the other hand, 45% or 19 species were unique in the pitch pine trunk samples and 50% (22 unique spe- cies) in oak trunk samples, suggesting that a greater proportion of species were using the trunk as an avenue to other habitats. The per- centage of species in 10% or more of the nat- ural habitats (Table 5) that occurred in the mailbox categories increased sequentially; in the unassigned category, 31%; in the seasonal category, 42%; and in the residential category, 83%. SUMMARY With the exception of the few species that deposited eggs and were subsequently ob- served as both juveniles and adults, none of the mailbox observations shed direct light on the manner in which various species arrived at the mailbox each year. Mailboxes are rela- tively isolated (see Fig. 2) and it is tempting to suggest that the presence of many species resulted from ballooning. Studies of spiders in agroecosystems, e.g.. Bishop & Riechert 1990, Rypstra & Carter 1995 and Young & Edwards 1990, strongly suggest that balloon- ing plays a significant role in the annual re- population of new habitats. Pitfall trap studies in various local habitats capture a surprising variety of species, typically dominated by old- er instars and adults. We suspect that the res- idential category (Table 2), including as it does species with relatively large numbers of early instars as well as adults later in the year, may be dominated by species that arrive ini- tially as a consequence of ballooning, and that the membership of the seasonal category (Ta- ble 3) is dominated by adults of species that entered “on foot”. This report examined the pattern of niche- spatial and temporal variations observed in the spiders present on mailboxes, and is ’ma- croecologicaF in nature {cf. Brown 1995). It was not feasible, given stringent time limita- 30 THE JOURNAL OF ARACHNOLOGY tions and other factors, for the mailman to sys- tematically collect and make observations. As a consequence, it is not possible to treat the mailbox data other than semiquantitatively. Nonetheless, these observations on a totally artificial habitat help to bring out emergent characteristics in spider niche selection and species assemblages. The number of unique and accidental species that parade through time and remain only briefly, initially suggests that colonization of the mailbox is almost a random process. However, the patterns ob- served are not as kaleidoscopic as it might first appear. There is a core assemblage on the mailbox represented by those species catego- rized as residential. This assemblage is peri- odically and consistently (and apparently suc- cessfully), challenged by other species at different and for shorter periods of time. This group in general is categorized as seasonal. In addition, there are yet other species, those in the unique and unassigned categories, which appear sporadically in time and in small num- bers, and are judged to be engaged in attempts to balloon or that are unsuccessful in gaining a foothold. To a certain extent, the data for the source habitats (Table 5) suggests that com- parable interactions may be involved. ACKNOWLEDGMENTS For assisting in species identifications, we wish to thank Vincent D. Roth, (Portal, Arizo- na); James H. Redner (Biosystematics Research Institute, Ottawa, Canada); and G.B. Edwards (Florida State Collections of Arthropods, Gainesville, Florida). We appreciated the many helpful suggestions made by the individuals who reviewed the manuscript for the journal. We are indebted to John B. Pearce (Northeast Fisheries Center, Woods Hole) for his comments and suggestions on the revised manuscript. LITERATURE CITED Bishop, L. & S.E. Riechert. 1990. Spider coloni- zation of agroecosystems: mode and source. En- viron. EntomoL, 19:1738-1745. Brown, J.H. 1995. Macroecology. Univ. Chicago Press, Chicago. Bultman, TL. & G.W. Uetz. 1983. Abundance and community structure of forest floor spiders fol- lowing litter manipulation. Oecologia, 55:34-41. Edwards, R.L. & E.H. Edwards. 1991. Spiders (Araneae) associated with rural delivery mail- boxes, Mashpee, Massachusetts. EntomoL News, 102:137-149. Edwards, R.L. 1993. Can the species richness of spiders be determined? Psyche, 100:185-208. Greenstone, M.H. 1984. Determinants of web spi- der species diversity; vegetation structural diver- sity vs. prey availability. Oecologia, 62:299-304. Gunnarsson, R. 1983. Winter mortality of spruce living spiders; effects of spider interactions and bird predation. Oikos, 40:226-233. Gunnarsson, R. 1992. Fractal dimension of plants and body size distribution of spiders. Funct. EcoL, 6:636-641. Hatley, C.L. & J.A. MacMahon. 1980. Spider com- munity organization; seasonal variation and the role of vegetation architecture. Environ. Ento- mol., 9:632-639. Heltsche, J.F. & N.E. Forrester. 1983. Estimating species richness using the jackknife procedure. Biometrics, 39:1-11. Hutchinson, G.E. 1957. Concluding remarks. Cold Spring Harbor Symp. Quant. Biol., 22:415-27. Millidge, A.F. 1981. The erigonine spiders of North America, Part 4. The genus Disembolus Chamberlin and Ivie (Araneae: Linyphiidae). J. Arachnol., 9:259-284. Moring, J.B. & K.W. Stewart. 1994. Habitat par- titioning by the wolf spider (Araneae, Lycosidae) guild in streamside and riparian vegetation zones of the Conejos River, Colorado. J. AranchnoL, 22:205-217. Riechert, S.E. & R.G. Gillespie. 1986. Habitat choice and utilization in web building spiders. Pp. 23-48. In, Spiders: Webs, Behavior and Evolution. (WA. Shear, ed.). Stanford Univ. Press, Stanford. Rushton, S.P. 1991. A discriminant analysis and logistic regression approach to the analysis of Walckenaeria habitat characteristics in grassland (Araneae: Linyphiidae). Bull. British Arachnol. Soc., 8:201-208. Rypstra, A.L. 1986. Web spiders in temperate and tropical forests; relative abundance and environ- mental correlates. American Midi. Nat., 115:42-51. Rypstra, A.L. & PE. Carter. 1995. The web-spider community of soybean agroecosystems in south- western Ohio. J. Arachnol., 23:135-144. Stratton, G.E., G.W Uetz & D.G. Dillery. 1978. A comparison of the spiders of three coniferous tree species. J. Arachnol., 6:219-226. Sundberg, I. & R. Gunnarsson. 1994. Spider abun- dance in relation to needle density in spruce. J. Arachnol., 22:190-194. Young, O.P. & G.B. Edwards. 1990. Spiders in United States field crops and their potential ef- fects on crop pests. J. Arachnol., 18:1-27. Manuscript received 31 August 1995, revised 15 May 1996. 1997. The Journal of Arachnology 25:25: 31-41 SPIDERS AND THEIR PREY IN MASSACHUSETTS CRANBERRY BOGS Carolyn J. Bardwell and Anne L. Averill: Department of Entomology, Femald Hall, University of Massachusetts, Amherst, Massachusetts 01003 USA ABSTRACT. Spiders from a total of 24 genera and eight families that possessed prey were collected using direct observation and sweep sampling during a survey of seven stands of wild (four sites) and abandoned (three sites) cranberry bogs in Massachusetts. Over all sites, 7009 spiders were inspected and 2.7% of all individuals possessed prey. At the wild bogs, Lycosidae and Araneidae were most commonly collected and at the abandoned bogs, Oxyopidae and Tetragnathidae were most common. A total of 11 orders of prey was observed and small Diptera (39.4% of total) (particularly Chironimidae), Collembola (18.6%), Homoptera (11.7%) (particularly Cicadellidae), and small Hymenoptera (9%) were the most common prey items. For all sites, three species of spider {Pardosa saxatilis Hentz (Lycosidae), Oxyopes salticus Hentz (Oxyopidae), and Tetragnatha laboriosa Hentz (Tetragnathidae)] represented 58% (109/ 188) of all specimens collected with prey. Sixty-seven percent of the prey recovered from P. saxatilis were Diptera or Collembola; another 20% were identified as Homoptera and Araneae. Collembola (35%) and Diptera (24%) were the dominant prey captured by O. salticus, and no predation on spiders by this species was observed. The majority of T. laboriosa with prey possessed chironomids (63%) or homop- terans (17%). Dvac® samples of vegetation, taken during the study to determine levels of the total potential prey, showed that the most abundant orders were Collembola, Diptera, and Araneae and Hymenoptera and that the number and type of prey taken by spiders fluctuated with the relative abundance of potential prey. Spiders (Arachnida, Araneae) are often the most ubiquitous and diverse insectivores in terrestrial ecosystems, exhibiting a variety of foraging strategies and prey preferences. Nu- merous surveys of spiders and their arthropod prey have been conducted in managed crop ecosystems, showing that spiders constitute a significant proportion of the predator guild (Young & Edwards 1990; Nyffeler et al. 1994); and in some systems, spiders are be- lieved to contribute to the biological control of arthropod pests (Riechert & Lockley 1984; Nyffeler & Benz 1987; Wise 1993). Unfortu- nately, in comparison to undisturbed or natu- ral environments, it is not surprising that it has also been shown that typical agroecosystems have considerably lower species diversity (Luczak 1975) and mean spider densities (Nyffeler et al., 1994). For example, in field crops, spider populations are adversely affect- ed by intensive disturbances, including culti- vation, mowing, harvesting and pesticide use. The cranberry, Vaccinium macrocarpon Ai- ton, is a perennial vine that is native to wet- lands in the northern regions of North Amer- ica. On wild and abandoned stands of cranberry, a great number of spiders can be found, but no investigations of their predatory role have been undertaken. In this article, we report the results of a study of spiders on wild and abandoned cranberry beds. We did not work on managed beds because we wished to determine the maximum levels of spider ac- tivity in the absence of extensive disturbances, particularly flooding and pesticide applica- tions, which are typical on the majority of commercial bogs. The main objectives of this study were to identify the most abundant spe- cies of spiders and the prey of these spiders in stands of cranberry. METHODS Study sites. — In 1992, surveys of spiders with prey items in their chelicerae were car- ried out in seven non-commercial cranberry bogs in eastern Massachusetts ranging in size from 0,2 to 1.2 ha. Bogs were classified as either “wild” or “abandoned” and were dom- inated by cranberry vines. The four wild bog sites were located in Sandwich (Sandy Neck), Truro (High Head), and Provincetown (Her- ring Cove and Mt. Ararat) in depressions be- 31 32 THE JOURNAL OF ARACHNOLOGY tween sand dunes. In addition to cranberry vines, other vegetation at these sites included Sphagnum moss, bayberry (Myrica pennsyl- vanica Mirbel), bog orchids (Habernaria spp.), round-leaved (Drosera rotundifoUa L.) and thread-leaved {Drosera filiformis Raf.) sundews, poison ivy (Rhus radicans L.), var- ious sedges, grasses and rushes, and other her- baceous and woody plants commonly found in undisturbed bog habitats in the region. The three abandoned bogs, located in Sandwich (Windmill) and Rochester (Mello 1 and Mello 2), were originally established for commercial cranberry production but were unmanaged for at least five years before this survey. Aban- doned bogs had thick mats of cranberry vines and Sphagnum moss interspersed with grass- es, brambles (Rubus spp.), poison ivy, small flowering shrubs, and saplings of the early successional tree species found in adjacent wooded habitats (including Acer rubrum L., Pinus strobus L., Populus spp., and Betula spp.). Bogs were considered to be composed of three overlapping strata: (1) ground surface beneath the vines, (2) dense cranberry vines, and (3) taller vegetation (composed of grasses, bayberry bushes, and tree saplings). As hunt- ing spiders generally forage on vegetation and web-builders trap prey from the air, we hy- pothesized that spiders within these strata would encounter and capture arthropod prey from the arthropod orders most commonly lo- cated there. Collection methods.— Surveys of spiders and their prey were conducted at all wild and abandoned bogs on alternate weeks between 1 June-28 August 1992. During this period, seven surveys were made at each site. Direct observation and sweep sampling to obtain spi- ders with prey were conducted at the sites be- tween 0930-1530 h, weather permitting. The first sampling method employed after arrival at a study site was direct observation. We di- vided each site into three parts of similar size, dependent on the total size of each site, using physical landmarks such as shrubs, trees, bog ditches, etc. Within each section, sampling was carried out by three observers who walked random paths for an hour. Bog vege- tation was visually searched for spiders, which were aspirated into clear, 30 ml cups and in- spected for the presence of prey in their mouthparts. If a spider possessed a prey item. alcohol was added to the cup to kill the spider; and both spider and prey were brought back to the laboratory for identification. If a spider did not possess a prey item, the inspection event was recorded on a hand-held counter and the spider was released. Next, 30 sets of five random sweeps were performed using a circular, 27.5 cm diameter cloth net, with each person sampling one-third of the bog. After five sweeps, the contents of the net were emptied into a light-colored dish pan; and spiders were quickly aspirated into cups and their mouth parts checked for prey. Spiders with and without prey were treated in the same manner as those captured during di- rect observation. Before leaving a site, 25 ran- domly-selected 0.20 m^ point samples of the arthropod fauna were obtained using a Dvac® (Dvac; Ventura, California) suction device (Dietrick 1961); and contents were placed in cyanide jars and transported back to the lab- oratory for identification. On several occa- sions, bogs were saturated with water, pre- venting use of the Dvac® and causing arthropod samples at some sites to be discon- tinuous. In addition to the seven-week surveys, two extra direct observation examinations were conducted between normally-scheduled sur- vey weeks at each of two abandoned bogs (Mello 1 and Mello 2) and two wild bogs (Herring Cove and Mt. Ararat). Direct obser- vations were performed in exactly the same manner as previously described; however, no sweep net or Dvac samples were taken. Identifications. — The first author identified all spider specimens collected with prey in the laboratory to genus, and species when possi- ble, using Kaston (1981). Voucher specimens preserved in 70% ethanol were sent to the American Museum of Natural History for confirmation and have been deposited in the University of Massachusetts entomology col- lection. Prey items removed from the mouth parts of spiders were identified to order. A few prey remains that were not identifiable were discarded and the capture event removed from the data record for the site where the spider with prey was collected. Arthropods from Dvac® samples were identified to order and preserved in 70% ethanol. Additional field trial. — In 1993, we com- pared the effectiveness of the direct observa- tion method we employed during 1992 with BARDWELL & A VERILL— SPIDERS AND PREY IN CRANBERRY BOGS 33 Table 1. — Percentage of spiders collected with prey in wild or abandoned cranberry bogs in 1992. Bog sites (ha) No. spiders inspected No. spiders with prey % spiders with prey Wild High Head (0.5) 947 37 3.9 Herring Cove (1.2) 962 23 2.4 Mt. Ararat (0.2) 968 36 3.7 Sandy Neck (0.8) 713 22 3.1 All wild bogs 3590 118 3.3 Abandoned Mello 1 (1.2) 982 23 2.3 Mello 2 (1.2) 1663 28 1.7 Windmill (1.2) 774 19 2.5 All abandoned bogs 3419 70 2.1 Total — all bog sites 7009 188 2.7 the “drunkard’s walk” (Southwood 1978) for capturing spiders with prey in the cranberry system. The latter method required the estab- lishment of a centered transect line at a site and use of a random numbers table to select discrete areas where direct observation was performed. Two people with observation ex- perience from the 1992 survey conducted the collection comparison. Once a point was se- lected, the spiders present within a 0.9 m ra- dius were individually captured and inspected during a 15 min period. A total of four indi- vidual points was selected by each observer during an hour. Three wild bogs from the 1992 study (Mt. Ararat, Herring Cove, and Sandy Neck) were selected for the comparison of the two sampling methods. Each site was surveyed weekly using both methods between 21 July- 12 August. A Wilcoxon signed rank test was used to compare the number of spi- ders inspected and the number of spiders col- lected with prey (Ott 1984). RESULTS Spiders collected with prey. — During the survey, 188 spiders with prey were collected. Twenty-four of the specimens (13%) were ob- tained from the additional direct observations performed at the Herring Cove, Mt. Ararat, Mello 1, and Mello 2 bogs. On average, 3.3% (118/3590) of the spiders inspected at the four wild bogs and 2.1% (70/3419) of the spiders inspected at the three abandoned bogs had prey items (Table 1). Sixty-one percent (115/188) of the spiders collected with prey were hunters and 39% (73/ Table 2. — Eamilies of spiders collected with prey at wild and abandoned cranberry bogs in 1992. Spider family Abandoned bogs Wild bogs All bogs n % n % n % Hunters (all) 47 67.1 68 57.6 115 61.2 Lycosidae 7 10.0 59 50.0 66 35.1 Oxyopidae 34 48.5 0 0 34 18.1 Salticidae 2 2.9 5 4.2 7 3.7 Thomisidae 2 2.9 4 3.4 6 3.2 Cubionidae 2 2.9 0 0 2 1.1 Web-builders (all) 23 32.9 50 42.4 73 38.8 Araneidae 4 5.7 2>1 31.4 41 21.8 Tetragnathidae 15 21.4 9 7.6 24 12.8 Linyphiidae 4 5.7 4 3.4 8 4.2 Total 70 100 118 100 188 100 34 THE JOURNAL OF ARACHNOLOGY Table 3. — Taxa of spiders and prey items collected in cranberry bogs in 1992. Order of prey: ARN = Araneae; COL = Collembola; DIP = Diptera; HOM = Homoptera; HYM = Hymenoptera; LEP = Lepidop- tera; OTH = Others (including: Orthoptera, Psocoptera, Coleoptera, Neuroptera and Acaii); TOT = Totals. Number of predation events Spider family ARN COL DIP HOM HYM LEP OTH TOT Web-building spiders Araneidae Argiope spp. 2 2 3 2 9 Acanthepeira spp. 1 1 Epeira spp. 1 1 Mangora gibberosa (Hentz) 4 2 3 1 3 13 Neoscona arabesca (Walckenaer) 7 1 3 11 Neoscona pratensis (Hentz) 2 1 3 Neoscona spp. 1 1 2 Singa spp. 1 1 All species Linyphiidae 1 0 18 5 10 1 6 41 Ceratinops spp. 1 1 Frontinella spp. 1 2 3 Helophora spp. 1 1 Neriene clathrata (Sundevall) 1 1 Neriene variabilis (Banks) 1 1 2 All species Tetragnathidae 0 1 4 0 2 0 1 8 Tetragnatha laboriosa 0 3 15 4 0 1 1 24 All web-building spiders Hunting spiders 1 4 37 9 12 2 8 73 Lycosidae Arctosa spp. 1 1 Lycos a spp. 1 1 Pardosa floridana (Banks) 1 2 2 1 6 Pardosa milvina (Hentz) 1 1 Pardosa modica (Blackwell) 1 1 Pardosa moesta (Banks) 1 3 1 5 Pardosa saxatilis (Hentz) 5 12 22 5 2 3 2 51 All species Oxyopidae 8 14 28 6 2 4 4 66 Oxyopes salticus (Hentz) 0 12 8 4 2 3 5 34 Salticidae Evarcha flammata (Clerck) 1 1 Habronattus spp. 1 1 Metaphidippus spp. 1 1 Paraphidippus spp. 3 1 4 All species 2 4 1 0 0 0 0 7 Clubionidae Clubiona spp. 1 1 Catianeira spp. 1 1 All species 0 1 ■ 0 1 0 0 0 2 Thomisidae Philodromus -spp. 1 1 Thanatus spp. 2, 1 1 4 Xysticus spp. 1 1 All species 2 0 0- 2 1 0 1 6 All hunting spiders 12 31 37 13 5 7 10 115 Totals 13 35 74 22 17 9 18 188 BARDWELL & A VERILL— SPIDERS AND PREY IN CRANBERRY BOGS 35 Table 4, — Seasonal variation in taxa of prey captured by spiders at four wild and three abandoned cranberry bogs in 1992. “ Percentages = the number of insects from a specific order ^ the total number of prey items captured by spiders during a month. Orders of prey captured by spiders” Month sampled (No. sampling events) Araneae Collem- bola Diptera Homop- tera Hymen- optera Lepi- doptera Other Total No. % No. % No. % No. % No. % No. % No. % No. Abandoned bogs June (11) 1 3.5 10 34.4 9 31.0 2 6.9 1 3.5 1 3.5 5 17.2 29 July (8) 0 0 5 22.7 9 40.9 3 13.6 1 4.6 2 9.1 2 9.1 22 August (6) 1 5.3 4 21.0 6 31.6 4 21.0 3 15.8 1 5.3 0 2.0 19 Wild bogs June (9) 5 25.0 7 35.0 5 25.0 2 10.0 0 0.0 0 0.0 1 5.0 20 July (13) 4 6.4 8 12.9 29 46.8 5 8.1 7 11.3 4 6.4 5 8.1 62 August (10) 2 5.5 1 2.8 16 44.4 6 16.7 5 13.9 1 2.8 5 13.9 36 All Bogs Season totals (57) 13 6.9 35 18.6 74 39.4 22 11.7 17 9.0 9 4.8 18 9.6 188 188) were web-builders (Table 2). Of the hunting spiders, 87% (100/115) were from the families Lycosidae (wolf spiders) and Oxy- opidae (lynx spiders). Spiders from the fami- lies Araneidae and Tetragnathidae (both orb weavers) made up 89% (65/73) of the web- builders with prey. Although eight families were represented in the survey, 88% (165/ 188) of all the predation events we witnessed involved lycosid, oxyopid, araneid, or tetrag- nathid spiders. The dominant families of hunters and web builders collected with prey differed between the wild and abandoned bogs. At the wild bogs, 81% (96/118) were lycosids and ara- neids, while at abandoned bogs, 70% (49/70) were oxyopids and tetragnathids. Lycosids and araneids were captured with prey and ob- served in high numbers at all of the wild bogs. All of the oxyopids captured with prey were from the abandoned Mello 1 and 2 bogs, al- though the presence of oxyopids at Windmill was noted during collection outings. In addi- tion, 13 of the 15 tetragnathids with prey from abandoned bogs were obtained at the Wind- mill bog. In total, 24 genera of spiders with arthropod prey from 1 1 orders were collected and iden- tified during the survey (Table 3). Three spe- cies (Pardosa saxatilis Hentz (Lycosidae), Oxyopes salticus Hentz (Oxyopidae), and Te- tragnatha laboriosa Hentz (Tetragnathidae)) represented 58% (109/188) of all specimens collected with prey. Sixty-seven percent (34/ 51) of the prey recovered from P. saxatilis were Diptera (22/51) or Collembola (12/51); another 20% (10/51) were identified as Ho- moptera (5/51) and Araneae (5/51). Collem- bolans (35%, 12/34) and dipterans(24%, 8/34) were the dominant prey captured by O. salti- cus, and no predation on spiders by this spe- cies was observed during thesurvey. The ma- jority of T. laboriosa with prey possessed chironomids (63%, 15/24) or homopterans (17%, 4/24). In addition to these three species, another 27% (51/188) of the spiders with prey were identified as various species of Pardosa, M angora (Araneidae), Neoscona (Araneidae), and Argiope (Araneidae). Thirty-nine percent (74/188) of the arthro- pods recovered from the chelicerae of all spi- ders captured with prey were dipterans. Small flies from the family Chironomidae represent- ed 51% (37/73) of all prey captured by web building species and 32%(37/l 15) captured by hunting species. Other orders frequently pos- sessed by the web-building spiders included the Hymenoptera (16%, 12/73) and Homop- tera (12%, 9/73). In addition to Diptera, the most common prey of spiders in the hunter guild were Collembola (27%, 31/115), Ho- moptera (11%, 13/115) and Araneae(10%, 12/115). During the months that sampling was con- ducted, fluctuations in the proportions of ar- thropod orders possessed as prey by spiders 36 THE JOURNAL OF ARACHNOLOGY Table 5. — Seasonal variation in potential prey at four wild and three abandoned cranberry bogs in 1992. Potential prey orders: Percentages = the number of insects from a specific order ^ the total number of potential prey collected during a month. Month sampled (No. sampling events) Araneae Collembola Diptera No. % No. % No. % Abandoned bogs June (8) 150 3.9 1534 40.3 1417 37.3 July (5) 392 7.7 2558 50.4 558 11.0 August (4) 942 30.0 1267 40.3 456 14.5 Wild bogs June (7) 99 2.7 2143 59.1 664 18.3 July (11) 798 11.4 1944 27.7 1801 25.7 August (7) 550 22.3 516 20.9 600 24.4 All Bogs Season totals (42) 2931 11.7 9962 39.6 5496 21.9 at the wild and abandoned bogs were evident (Table 4). Between June- August at the wild and abandoned bogs, the proportion of ho- mopteran and hymenopteran prey captured by spiders increased, while the proportion of col- lembolan prey taken decreased. During the same interval, the proportion of the total prey from the orders Lepidoptera and Diptera was greatest in July. Araneid prey items comprised a larger proportion of the total prey taken by spiders at the wild bogs(from 6-25%) than at the abandoned bogs (from 0-5.3%) through- out the study. Potential prey. — Dvac® samples taken dur- ing June, July, and August showed fluctua- tions in the abundance of potential prey (ar- thropods available to foraging spiders). The number of arthropods collected per sample was greatest during July at both wild and abandoned bogs (Table 5). Collembola were the most abundant poten- tial prey at the abandoned sites, comprising 40-50% of all arthropods collected each month. In addition, the emergence of chiron- omids in June, adult Lepidoptera in July, and oxyopid spiderlings in August was reflected in the composition of the samples from the abandoned bogs. At the wild sites, the proportion of Collem- bola steadily declined from 59% (2143/3628) of the total potential prey in June, to just 21% (516/2464) in August. During July, increased numbers of arthropods from the orders Ara- neae, Diptera, Homoptera, and Hymenoptera were evident in the samples from the wild sites. Of the total potential prey present in Dvac® samples from all wild and abandoned sites throughout the survey, the most abundant orders were (in descending order) Collembola, Diptera, Araneae and Hymenoptera. Comparison of collection methods. — Of the two collection methods we employed in the cranberry system, direct observation was generally more effective for capturing spiders with prey than sweep netting. Although the mean number of spiders inspected using the two methods was similar over all seven sites surveyed, the mean number of spiders col- lected with prey was greater using the direct observation method (P = 0.0001, Wilcoxon signed rank test) (Table 6). During the field trial conducted in 1993, the protocol for direct observation used in the 1992 survey resulted in both a greater mean number of spiders inspected (P = 0.001, Wil- coxon signed rank test) and collected with prey (P = 0.003, Wilcoxon signed rank test) than the “drunkard’s walk” method (Table 7). DISCUSSION Spiders collected with prey.— Over all sites, approximately 2.7% (188/7009) of the spiders that we inspected possessed prey. In the literature, the percentage of hunting spi- ders collected while feeding ranges from 1.4- 8.3% (Nyffeler et al. 1987b, 1989; Young 1989), while <10%-12% has been reported for web-builders (Nyffeler et al. 1989; LeSar BARDWELL & A VERILL— SPIDERS AND PREY IN CRANBERRY BOGS 37 Table 5. — Extended. Homoptera Hymenoptera Lepidoptera Other Total No. No. % No. % No. % No. % 136 3.6 448 11.8 33 0.9 83 2.2 3801 204 4.0 565 11.1 633 12.5 170 3.3 5080 80 2.6 261 8.3 72 2.3 64 2.0 3142 228 6.3 357 9.8 21 0.6 116 3.2 3628 786 11.2 932 13.3 190 2.7 557 8.0 7008 166 6.7 295 12.0 221 9.0 116 4.7 2464 1600 6.4 2858 11.4 1170 4.6 1106 4.4 25 123 & Unzicker 1978). Although collecting tech= nique, vegetational architecture, spider spe- cies, potential prey and several other factors varied among the studies, the average percent of spiders with prey in unmanaged cranberry systems falls within the range of that found in other systems. Of the total spider fauna found in field crops grown in the United States, 56% are es- timated to be hunting species and 44% web- building species (Young & Edwards 1990). Surveys performed in alfalfa, peanuts, rice, and cotton cite percentages ranging from 42- 93 for hunting spiders and 17-58 for web- builders (Wheeler 1973; Agnew & Smith 1989; Heiss & Meisch 1985; Whitcomb et al. 1963). We found the proportions of spider types collected with prey in cranberries to be similar to these other crops, i.e., 61% were hunting species and 39% web-building spe- cies. Though the diversity of species was not determined, it is likely that these values reflect the general composition of spider types pres- ent on cranberry bogs. The feeding trends of spiders collected with prey at wild and abandoned cranberry bogs indicate that many of the web-building and hunting species present have a varied diet that is dominated by adult dipterans. Of the 188 spiders collected with prey, 51% of all web- builders and 32% of all hunters possessed dip- teran prey. Relatively high proportions of Dip- tera (up to 77.8% of all prey captured) have also been reported in the diets of many web- building and hunting spiders in soybean, cot- ton, wheat field, alfalfa, and grassland ecosys- tems (Nyffeler et al. 1994). In general, spiders collected with prey in cranberries possessed arthropods of types located in the microhabitat where the spider’s foraging activity was con- centrated; hunters possessed prey from the ground and the cranberry vine strata, web- builders prey from the vine and tall vegetation strata. Prey data for hunting spiders in many sys- tems indicate that although a variety of ar- thropod taxa are accepted, the groups most commonly captured include Collembola, Dip- tera, Heteroptera and Araneae (Edgar 1969; Hallander 1970; Yeargan 1975; Nyffeler et al. 1992, 1994). In addition to dipterans, most hunting spiders in cranberries possessed prey from orders located primarily on the ground or in the vines of bog, specifically, Collem- bola (27%), leafhoppers (11%), and immature spiders (10%). The species of hunting spiders most fre- quently collected with prey in cranberry were Pardosa saxatilis Hentz and Oxyopes salticus Hentz. P. saxatilis was collected with a wide range of prey that was dominated by Diptera and Collembola, but occasionally included species of leafhoppers that vector cranberry false blossom disease and Lepidoptera whose larvae are foliar pests in the cranberry system. Yeargan (1975) observed that, despite an abundance of lepidopterans in alfalfa, the diet of the lycosid Pardosa ramulosa McCook 38 THE JOURNAL OF ARACHNOLOGY Table 6. — Comparison of the mean number of spiders inspected and collected with prey using direct observation and sweep net methods in 1992. " Spiders with prey collected during additional visits to sites were not used in comparison calculations, n = 2\ hours for direct observations and n — 2\0 sets of five sweeps for sweepnet samples at each bog. Significantly more spiders with prey were collected using the direct observation method {P = 0.0001, Wilcoxon signed rank test). Mean number ± SE Collected Bog Method used"* Inspected with prey High Head Direct observation 25.0 ± 2.7 1.6 ± 0.4 Sweepnetting 20.0 ± 2.1 0.1 ± 0.1 Herring Cove Direct observation 24.7 ± 1.9 0.8 ± 0.2 Sweepnetting 19.0 ± 2.4 0 Mt. Ararat Direct observation 24.2 ± 2.2 1.2 ± 0.3 Sweepnetting 14.7 ± 1.4 0 Sandy Neck Direct observation 26.2 ± 3.9 1.0 ± 0.2 Sweepnetting 7.7 ± 1.3 0 Mello 1 Direct observation 21.7 ± 2.2 0.7 ± 0.2 Sweepnetting 19.0 ± 3.2 0.2 ± 0.2 Mello 2 Direct observation 32.8 ± 4.0 0.6 ± 0.2 Sweepnetting 78.7 ± 37.9 0.4 ± 0.1 Windmill Direct observation 17.8 ± 2.3 0.7 ± 0.2 Sweepnetting 19.3 ± 3.2 0.2 ± 0.1 All bogs Direct observation 24.6 ± 1.1 1.0 ± O.D Sweepnetting 25.5 ± 5.7 0.1 ± 0.0 consisted primarily of prey from the orders these factors may have affected the prey se- Homoptera, Diptera, and Araneae. Yeargan lection we observed for P. saxatilis. concluded that the predation exhibited by P. ramulosa may have been due to the rarity of encounters with Lepidoptera, which were lo- cated in the foliage above areas where the spi- ders most often foraged, and to attractiveness of the sudden movements often made by ho- mopteran and dipteran prey. Although lepi- doptera were scarce in the bogs we sampled. Predation of spiders by oxyopids has been reported in several surveys conducted in cot- ton and wooley croton, Croton capitatus Mi- chaux, in Texas (Nyffeler et al. 1987a, 1987b, 1992). However, Lockley & Young (1987) noted a conspicuous lack of spiders possessed as prey by O. salticus in cotton in Mississippi, USA. Our data on the feeding behavior of O. Table 7. — Comparison of the mean number of spiders inspected and collected with prey using the “drunkard’s walk” and 1992 direct observation methods. ^ n = 8 h for each method at each site. Signif- icantly more spiders were inspected*’ (P = 0.001, Wilcoxon signed rank) and collected with prey'’ (P = 0.003, Wilcoxon signed rank) using the direct observation technique. Mean number ± SE Collected Bog Method used"* Inspected with prey Herring Cove Drunkard’s walk 9.5 ± 3.0 0.3 ± 0.3 Direct observation 54.8 ± 7.3 2.8 ± 0.5 Mt. Ararat Drunkard’s walk 22.8 ± 2.1 0.5 ± 0.3 Direct observation 50.0 ± 5.6 3.5 ± 1.5 Sandy Neck Drunkard’s walk 8.5 ± 3.5 0.5 ± 0.5 Direct observation 40.5 ± 7.2 3.3 ± 2.0 All bogs Drunkard’s walk 13.6 ± 2.5 0.4 ± 0.2 Direct observation 48.4 ± 4.0*’ 3.2 ± 0.8^^ BARDWELL & AVERILL— SPIDERS AND PREY IN CRANBERRY BOGS 39 salticus in abandoned cranberry bogs con- curred with the latter findings for unknown reason(s), but which may have involved the availability of more easily captured prey items. Studies of orb-weaving spiders (Araneidae and Tetragnathidae) in temperate regions have shown that most species capture Homoptera, Diptera, and small parasitic Hymenoptera in their webs (Nentwig 1987; Culin & Yeargan 1982; Provencher & Coderre 1987). In addi- tion, large orb- weavers (Argiope spp.) may feed on aculeate Hymenoptera, grasshoppers, and various other “difficult” prey (Nentwig 1985; Nyffeler et al. 1987c, 1989, 1991). Our data show that orb spiders capture winged prey, predominantly Diptera, Hymenoptera and Homoptera, flying between cranberry vines and tall vegetation in bogs. Though un- common in our study, several large-bodied Hymenoptera and Orthoptera were captured by females of the genus Argiope in late Au- gust as the spiders approached maturity. The majority of prey captured by species of sheet- web spiders (Linyphiidae) were from the same orders as those captured by orb- weavers. The web-builder Tetragnatha laboriosa Hentz, one of the most frequently occurring spider species in field crops in the USA (Young & Edwards 1990), has been shown to commonly capture heteropteran and dipteran prey in cotton and soybean systems (LeSar & Unzicker 1978; Culin & Yeargan 1982; Nyf- feler et al. 1989). In our survey of unmanaged cranberry bogs, T. laboriosa was the species of web-building spider most frequently ob- served with prey. The orders of prey pos- sessed most often, Diptera and Homoptera, were consistent with the dominant arthropod groups reported for this species in the agri- cultural systems previously mentioned. Spiders are considered by many to be gen- eralist predators that capture the prey species that are most abundant, and thus most often encountered, in their environment (Turnbull 1960; Riechert & Lockley 1984; Wise 1993). Comparison between the proportion of prey captured by spiders and frequencies of poten- tial prey in cranberry bogs indicates that spi- der predation is influenced by prey abundance (Tables 4 and 5). Although the numbers of spiders collected with prey were low, the number and type of prey possessed by spiders fluctuated with the relative abundance of po- tential prey as captured in Dvac® samples, for most of the arthropod orders present in the system. Foliage-feeding lepidopteran larvae and adult cranberry weevils (Anthonomus muscu- lus Say) are the primary pest insects found in commercial cranberry bogs. Of the 13 spiders captured with lepidopteran prey in non-com- mercial bogs, five of the prey items were lar- vae. Two spiders were collected with coleop- teran prey items during the study; however, neither was a cranberry weevil. In sum, our data indicate that few spiders in non-commer- cial cranberry bogs capture pest insects such as lepidopteran larvae or weevils. This sug- gests that spiders with similar predation be- havior in commercial bogs may not have a very high impact on insect pests, particularly low density populations such as were present in the non-commercial systems. Comparison of collection methods. — Over the seven-week period of this study, di- rect observation was more effective than sweepnetting in collecting spiders with prey. Spiders collected using sweepnets were often damaged and rarely possessed prey. Both in- jury to the spiders and absence of prey was most likely the result of the sweeping motion and tumbling contents of the net. Under such conditions, it is likely that spiders entering a sweepnet with prey in their mouth parts re- sponded by releasing their prey. Prey may have also been released by spiders as the sweep samples were emptied into the dish pan and inspected. Mean numbers of spiders inspected and col- lected with prey during 1993 show that the direct observation method used in 1992 was also more effective than the “drunkard’s walk” method in the cranberry system, de- spite the successful use of the latter method in other systems (S.E. Riechert, pers. comm.). Unlike many row crops, the cranberry bogs we surveyed were covered in dense layers of vine with little exposed substrate. Hunting spi- ders were only visible when they were resting or actively moving on the uppermost layer of cranberry vine. Web-building spiders often positioned themselves in grasses and shrubs above the vines where they were visible to observers. Although they were easily spotted, there were not many present in any given area of bog. Given such circumstances, the prob- ability of locating a hunting or web-building 40 THE JOURNAL OF ARACHNOLOGY spider may have been improved by using the direct observation method because of the in- creased proportion of bog area searched by experimenters. ACKNOWLEDGMENTS We gratefully acknowledge Jessica B. Dunn’s invaluable assistance in data collec- tion. Cape Cod National Seashore and Sandy Neck Governing Board provided access to study sites. This study was supported by USDA/EPA project ANE 91.2, Massachusetts Agricultural Experiment Station Project H-673 and Cape Cod Cranberry Growers’ As- sociation. LITERATURE CITED Agnew, C.W. & J.W. Smith, Jr. 1989. Ecology of spiders (Araneae) in a peanut agroecosystem. En- viron. Entomol., 8:30-42. Culin, J.D. & K.V. Yeargan. 1982. Feeding behav- ior and prey of Neoscona arabesca (Araneae: Araneidae) and Tetragnatha laboriosa (Araneae: Tetragnathidae) in soybean fields. Entomophaga, 27:417-424. Edgar, W.D. 1969. Prey and predators of the wolf spider Lycosa lugubhs. J. Zool., London, 159: 405-411. Hallander, H. 1970. Prey, cannibalism and micro habitat selection in the wolf spiders Pardosa che- lata O.E Muller and P. pullata Clerck. Oikos, 21:337-340. Heiss, J.S. & M.V. Meisch. 1985. Spiders (Ara- neae) associated with rice in Arkansas with notes on species compositions of populations. South- west. Nat., 30:119-127. Kaston, B.J. 1981. Spiders of Connecticut. State Geol. Nat. Hist. Surv. Connecticut. Bull. 70, Hartford, Connecticut. 1020 pp. LeSar, C.D. & J.D. Unzicker. 1978. Life history, habits, and prey preferences of Tetragnatha la- boriosa (Araneae: Tetragnathidae). Environ. En- tomol., 7:879-884. Luczak, J. 1975. Spider communities of the crop fields. Polish Ecol. Stud. 1: 93-110. Lockley, T.C. & O.P Young. 1987. Prey of the striped lynx spider Oxyopes salticus (Araneae, Oxyopidae), on cotton in the delta area of Mis- sissippi. J. ArachnoL, 14:395-397. Nentwig, W. 1985. Prey analysis of four species of tropical orb- weaving spiders (Araneae: Aranei- dae) and a comparison with araneids of the tem- perate zone. Oecologia, 66:580-594. Nentwig, W. 1987. The prey of spiders. Pp. 249- 263, In, Ecophysiology of Spiders. (W. Nentwig, ed.). Springer- Verlag, Berlin, New York. Nyffeler, M. & G. Benz. 1987. Spiders in natural pest control: A review. J. Appl. Ent., 103:321- 339. Nyffeler, M. & R.G. Breene. 1991. Impact of pre- dation upon honey bees (Hymenoptera, Apidae), by orb-weaving spiders (Araneae, Araneidae and Tetragnathidae) in grassland ecosystems. J. Appl. Ent., 111:179-189. Nyffeler, M., D.A. Dean & W.L. Sterling. 1987a. Predation by green lynx spider, Peucetia viridans (Araneae: Oxyopidae), inhabiting cotton and woolly croton plants in east Texas. Environ. En- tomol., 16:355-359. Nyffeler, M., D.A. Dean & W.L. Sterling. 1987b. Evaluation of the importance of the striped lynx spider, Oxyopes salticus (Araneae: Oxyopidae), as a predator in Texas cotton. Environ. Entomol., 16:1114-1123. Nyffeler, M., D.A. Dean & W.L. Sterling. 1987c. Feeding ecology of the orb-weaving spider Ar- giope aurantia (Araneae: Araneidae) in a cotton agroecosystem. Entomophaga, 32:367-375. Nyffeler, M., D.A. Dean & W.L. Sterling. 1989. Prey selection and predatory importance of orb- weaving spiders (Araneae: Araneidae, Ulobori- dae) in Texas cotton. Environ. Entomol., 18:373- 380. Nyffeler, M., W.L. Sterling & D.A. Dean. 1992. Impact of the striped lynx spider (Araneae: Ox- yopidae) and other natural enemies on the cotton fleahopper (Hemiptera: Miridae) in Texas cotton. Environ. Ent., 21:1178-1188. Nyffeler, M., W.L. Sterling & D.A. Dean. 1994. How spiders make a living. Environ. Ent., 23: 1357-1367. Ott, L. 1984. An introduction to statistical methods and data analysis. PWS Publishers, Boston. 775 pp. Provencher, L. & D. Coderre. 1987. Functional re- sponses and switching of Tetragnatha laboriosa Hentz (Araneae: Tetragnathidae) and Clubiona pikei Gertsh (Araneae: Clubionidae) for the aphids Rhopalosiphurn maidis (Fitch) and Rho- palosiphum padi (L.) (Homoptera: Aphididae). Environ. Entomol., 16:1305-1309. Riechert, S.E. & T Lockley. 1984. Spiders as bi- ological control agents. Ann. Rev. Entomol., 29: 299-320. Southwood, T.R.E. 1978. Ecological Methods. Chapman & Hall; London, New York. 524 pp. Turnbull, A.L. 1960. The prey of the spider Liny- phia triangularis (Clerck) (Araneae, Linyphi- idae). Canadian J. Zool., 38:859-873. Wheeler, A.G., Jr. 1973. Studies on the arthropod fauna of alfalfa: V. Spiders (Araneidae). Cana- dian Ent., 105:425-432. Whitcomb, W.H., H. Exline & R.C. Hunter. 1963. Spiders of the Arkansas cotton field. Ann. En- tomol. Soc. America, 56:653-660. BARDWELL & A VERILL— SPIDERS AND PREY IN CRANBERRY BOGS 41 Wise, D.H. 1993. Spiders in Ecological Webs. Cambridge Univ. Press, Cambridge. 328 pp. Yeargan, K.V. 1975. Prey and periodicity of Par- dosa ramulosa (McCook) in alfalfa. Environ. En- tomoL, 4:137-141. Young, O.P. 1989. Field observations of predation by Phidippus audax (Araneae: Salticidae) on ar- thropods associated with cotton. J. Entomol. Sci., 24:266-273. Young, O.P. & G.B. Edwards. 1990. Spiders in United States field crops and their potential effect on crop pests. J. ArachnoL, 18:1-27. Manuscript received 12 June 1995, revised 23 July 1996. 1997. The Journal of Arachnology 25:42-48 ESTRUCTURA OCULAR DE SELENOPS COCHELETI (ARANEAE, SELENOPIDAE) Jose Antonio Corronca: CONICET-Fundacion M. Lillo-INSUE Fac. de Cs. Nat. e Inst. Miguel Lillo 251. (4000) S. M. de Tucuman, Argentina Hector R. Teran: Institute de Morfologia Animal, Fundacion Miguel Lillo. Miguel Lillo 251. (4000) S. M. de Tucuman, Argentina ABSTRACT. Selenops cocheleti Simon 1880 (Selenopidae, Araneae) have their eight eyes arranged in two rows, the first formed by six eyes and the second row by only two. The first row of eyes is formed by two anterior lateral (ALE), two posterior median (PME) and two anterior median eyes (AME). Two posterior lateral eyes (PLE) point backwards and form the second row of eyes. This study used the histological staining technique of Hematoxylin-Eosin and Mallory-Azan-Heindenhain. The specimens were fixed with Bouin, using n-Butyl alcohol as an intermediary for embedding in Paraplast. Frontal, transversal and sagittal sections (6-10 ixm) were made. The cornea and lens of all the eyes are cuticular and laminar. Behind the lens there is a layer of cone cells with projections toward the lens. The direct eyes have two cellular types at the rhabdoma. They support the pigmented cells and the sensitive cells of the retina. In the indirect eyes three cellular types are found: sensitive cells of the retina, pigmented support cells and unpigmented support cells. These eyes have a tapetum (layer which reflects the light) below the pigmented layer of the ocular cup. RESUMEN. Selenops cocheleti Simon 1880 (Selenopidae, Araneae) posee sus ocho ojos dispuestos en dos filas, la primera formada por seis ojos y la segunda por solo dos. La fila anterior de ojos esta constitmda por dos ojos laterales anteriores (OLA), dos medios posteriores (OMP) y dos medios anteriores (OMA). Los ojos laterales posteriores (OLP) se dirigen hacia atras y forman la segunda fila de ojos. Los OMA son ojos principales y los restantes secundarios. Para este estudio se utilizaron tecnicas histomorfologicas con la coloracion de Hematoxilina-Eosina y Mallory-Azan-Heidenhain. Los ejemplares fueron fijados con Bouin utilizando como intermediario alcohol n-Butflico para su inclusion en Paraplast. Se realizaron cortes a 6— 10 |jLm, en seccion frontal, transversal y sagital. La cornea y la lente de todos los ojos son cuticular y laminar; detras de la lente se extiende una capa uniestratificada de celulas cono con proyecciones hacia la lente. Los ojos principales tienen dos tipos celulares en el rabdoma; las celulas de soporte pigmentadas y las celulas sensoriales de la retina. En los ojos secundarios se reconocieron tres tipos celulares; las celulas de la retina, las celulas de soporte pigmentadas y las de soporte no pigmentadas. En estos ultimos se encuentra tambien un tapete (capa que refleja la luz) delante de la capa pigmentada de la copa ocular. La disposicion de los ojos de los selenopi- dos es caractenstica y esta relacionada con la forma achatada del cuerpo. Los ojos se dis- ponen en una fila anterior formada por seis ojos y la posterior por solo dos dirigidos hacia atras, correspondientes a los ojos laterales posteriores (OLP). Los ojos medios anteriores (OMA) y medios posteriores (OMP) que en Selenops se disponen proximos entre si y de manera recta o levemente recurvada en la por- cion central de la region cefalica (Fig. 1). Los ojos laterales anteriores (OLA), ubicados la- tero-extemamente y separados de los ojos me- dios, carecen de anillo pigmentado, son per- lados y considerados como los unicos ojos noctumos de estas aranas (Simon 1897; FO. Pickard-Cambridge 1900). Homann (1971) describe la estructura de los ojos de Seleno- pidae y la compara con los de Sparassidae. Desde ese ano hasta la fecha no se ha vuelto a estudiar la estructura de los ojos de Selen- opidae y se desconocen muchos aspectos re- lacionados a la vision y la estructura general del aparato optico de estas aranas y las rela- clones que pudieran existir con otras familias. Este es el primer trabajo de una serie de estudio sobre la anatomfa ocular de familias relacionadas con Selenopidae. En este trabajo 42 CORRONCA & TERAN— ESTRUCTURA OCULAR DE SELENOPS COCHELETI 43 Figuras 1-4. — Selenops cocheleti Simon. 1. disposicion de los ojos; 2-4. Ojo medio anterior (OMA); 2. Seccion frontal (160X); 3. Seccion longitudinal (250X); 4. Seccion frontal mostrando rabdomeros y nucleos de las celulas sensoriales de la retina (lOOOX). Abreviaturas: OMA, ojos medios anteriores; OLA, ojos laterales anteriores; OMP, ojos medios posteriores; OLP, ojos laterales posteriores; c, cornea; 1, lente; cc, celulas cono; nc, nucleo de celula cono; no, nervio optico; np, nucleo de celula de soporte pigmentada; ns, nucleo de celula sensorial; rh, rabdomero. se describe los ojos de Selenops cocheleti Simon. METODOS Se utilizaron seis ejemplares, tres subadultos (ld2$) y tres adultos (IS 29), de Selenops cocheleti Simon capturados bajo corteza de Eu- caliptus sp. en el Departamento Capital, Santi- ago del Estero, Argentina, incluidos en la co- leccion de la Fundacion Miguel Lillo (FML N° 02100 y 02101). Los especfmenes voucher per- manecen depositados en la histoteca de la Co- leccion de Aracnidos de la Fundacion Miguel Lillo, cuyos lotes corresponden a los numeros antes mencionados. Se realizaron disecciones de las regiones cefalicas las que fueron fijadas en Bouin. El material se incluyo en Paraplast y se utilizo como intermediario en la deshidratacion alcohol n-Butflico. Se efectuaron cortes seriados de 6 y 10 jam siguiendo los pianos sagital, frontal y trans- versal. La preparaciones fueron tenidas con Hematoxilina-Eosina y Mallory-Azan-Hei- denhain. RESULTADOS Ojos medios anteriores (OMA). — Cor- nea: Cuticular, laminar con superficie externa 44 THE JOURNAL OF ARACHNOLOGY ondulada. Las ondulaciones se continuan en una estna transversal que la atraviesa en toda su espesor, dando un aspecto facetado (Fig. 2). Lente: Se ubica por debajo de la cornea, es una estructura laminar, biconvexa, con la con- vexidad mayor hacia la parte interna del ojo (Fig. 2). Las laminas son mas anchas que las de la cornea y acompanan la curvatura de la lente. Celulas cono (Eakin & Brandenburger 1971): se ubican por debajo de la lente, po- seen nucleos basales ovoides, con cromatina condensada (Figs. 2, 4). Se disponen en un solo estrato con proyecciones hacia la lente. Las celulas cono descansan sobre una delgada membrana basal que las separa de la retina (Figs. 2-4). Retina: Semilunar, formada por dos tipos celulares, las celulas de soporte pigmentadas y las sensoriales de la retina. Las celulas de soporte pigmentadas, con nucleos con cro- matina granular dispuesta homogeneamente, contienen numerosos granulos de pigmento concentrandose anteriormente formando una capa pigmentada oscura (Fig. 3). Los nucleos de estas celulas se disponen en diferentes pianos, dando un aspecto desordenado (Fig. 4). Hacia la parte profunda, el niimero de granulos de pigmento disminuye formando bandas ampliamente separadas por el citoplas- ma de las celulas sensoriales de la retina (Fig. 3). Cada celula sensorial de la retina esta for- mada por la zona distal, ocupada por los rab- domeros (Figs. 3, 4), inmediatamente por de- bajo de la membrana basal; un segmento intermedio que atraviesa la capa oscura y clara de las celulas de soporte pigmentadas y la por- cion nuclear (Fig. 3) donde la celula aumenta su volumen. Los nucleos de las celulas sen- soriales con cromatina granular distribuida homogeneamente (Fig. 4), se disponen peri- fericamente en la base de la copa ocular, em- itiendo prolongaciones intemas que forman el nervio optico. En un corte transversal del ojo los nucleos de las celulas sensoriales se dis- ponen en forma sublineal con las prolonga- ciones sensoriales a modo de paquete. Por la estructura antes mencionada estos ojos se de- nominan de vision directa. Ojos medios y laterales posteriores (OMP y OLP). — Cornea similar a OMA. Lente biconvexa, de forma ojival y celulas cono semej antes a las descritas para los OMA. Retina esta formada por tres tipos celulares. las celulas sensoriales de la retina (Figs. 5, 8), las celulas de soporte pigmentadas y las de soporte no pigmentadas (Figs. 5, 6). Existe di- ferencia en la estructura retiniana en los OMP y OLP Los primeros poseen una estructura retiniana subrectangular (Fig. 8) mientras que los OLP en forma de copa (Fig. 5). Ojos laterales posteriores (OLP). — Los nucleos de las celulas sensoriales, con cro- matina distribuida homogeneamente (Fig. 6), se ubican inmediatamente por debajo de la membrana basal y otros se agrupan lateral- mente entre la capa de pigmento hasta el ver- tice inferior. De los somas de las celulas sen- soriales se prolongan el segmento intermedio continuandose en los rabdomeros dispuestos paralelamente (Fig. 7) cuyas prolongaciones atraviesan el reducido tapete de tipo “RT” se- gun Homann (1971) (Fig. 7). Las celulas de soporte no pigmentadas son celulas gigantes y en numero reducido (Fig. 5). El niicleo con cromatina granular, en algunos casos dispuesta perifericamente, y citoplasma claro, abundan- te y con proyecciones entre las fibras de las celulas sensoriales, a menudo acompana a los rabdomeros llegando hasta la copa pigmenta- da (Figs. 5, 6). Las celulas de soporte pig- mentadas poseen un citoplasma con mayor cantidad de granulos de pigmento cerca de los rabdomeros y forman bandas convergentes ha- cia el nervio optico, disminuyendo la concen- tracion de granulos de pigmento (Fig. 5). Los nucleos son dificiles de observar debido a la gran cantidad de pigmento. Ojos medios posteriores (OMP). — Las celulas sensoriales de la retina dispuestas en por lo menos dos capas (Fig. 8), con los nu- cleos llegando hasta el vertice intemo de la capa pigmentada (Fig. 9). Los rabdomeros se distribuyen paralelamente formando una faja subrectangular (Fig. 8). Las celulas de soporte no pigmentadas son grandes, con nucleos con cromatina granular, y poseen abundante cito- plasma claro con prolongaciones entre las fi- bras de las celulas sensoriales de la retina, pero no atraviesan los rabdomeros (Fig. 9). Las celulas de soporte pigmentadas poseen mayor concentracion de granulos de pigmento en la zona de contact© con los rabdomeros y en la periferia de la copa ocular, disponien- dose entre estas regiones en bandas de menor numero de granulos de pigmento. Ojos laterales anteriores (OLA). — Cor- nea: Cuticular, laminar, con la superficie ex- CORRONCA & TERAN— ESTRUCTURA OCULAR DE SELENOPS COCHELETI 45 Figuras 5-8. — Selenops cocheleti Simon. 5-7, Ojo lateral posterior (OLP). 5. Seccion longitudinal (250X); 6. Detalle de celula de soporte no pigmentada (lOOOX); 7. Detalle de tapete y rabdomeros (lOOOX). 8-9. Ojo medio posterior (OMP); 8. Seccion longitudinal mostrando rabdomeros (250X); Abrev- iaturas: nc, nucleo de celula cono; no, nervio optico; ns, nucleo de celula sensorial; rh, rabdomero; sn. celula de soporte no pigmentada; t, tapete. tema con aproximadamente 4-5 placas imbri- cadas separadas unas de otras por estrias que atraviesan todo el espesor de la lente (Fig. 10). Lente: Laminar, ligeramente ojival (Figs. 10, 12). Celulas cono: Se ubican por debajo de la lente, sus nucleos son ovales, con cro- matina muy condensada y se disponen en un solo estrato, no observandose proyecciones hacia las lentes como en los restantes ojos (Fig. 10). Retina: Se observan tres tipos celulares, celulas sensoriales de la retina, celulas de so- porte no pigmentadas y celulas de soporte pig- mentadas. Los nucleos de las celulas sensoriales 46 THE JOURNAL OF ARACHNOLOGY Figuras 9-12. — Selenops cocheleti Simon. 9. Seccion longitudinal mostrando celulas de soporte no pigmentadas (400 X). 10. Ojo lateral anterior (OLA) mostrando forma de lente y superficie externa de la cornea (lOOOX); 11. Ojo lateral posterior, en seccion frontal, mostrando forma de la cornea y lente (160X); 12. OLA, seccion longitudinal mostrando disposicion de nucleos de celulas sensoriales de la retina y celulas de soporte no pigmentadas (400 X). Abreviaturas; ant, anterior; c, cornea; 1, lente; post, posterior; nc, nucleo de celula cono; no, nervio optico; ns, nucleo de celula sensorial; sn, celula de soporte no pigmentada; t, tapete. se ubican en posicion latero-posterior-extema del ojo (Fig. 12), emitiendo los segmentos in- termedios hacia el centro de la copa ocular donde se ordenan los rabdomeros, en numero escaso, dispuestos de manera semilunar y con prolongaciones que atraviesan el tapete (Fig. 12) de tipo “canoe” (KT, segun Homann 1971). Las celulas de soporte no pigmentadas son gigantes, con cromatina granular dispues- ta homogeneamente, citoplasma claro, abun- dante (Fig. 12) y con gruesas proyecciones ha- cia los somas de las celulas sensoriales de la retina. Las celulas de soporte pigmentadas presentan los granulos de pigmento concen- trados de manera lineal detras del tapete y per= ifericamente a la copa ocular, especialmente en la porcion latero-posterior-extema; entre estas regiones el pigmento se distribuye de- sordenadamente, sin formar bandas paralelas como en los restantes ojos. El pigmento con- verge hacia el nervio optico al cual acompana en gran parte de su extension (Fig. 12). Los nucleos de las celulas de soporte pigmentadas son grandes, no pudiendose determinar la dis- CORRONCA & TERAN— ESTRUCTURA OCULAR DE SELENOPS COCHELETI 47 posicion de la cromatina por estar enmascar- ada por los granules de pigmento. DISCUSION Tipos celulares. — En los ojos de Selenops cocheleti Simon se observaron diferentes tipos celulares. Los OMA (principales) presentan dos tipos celulares, celulas de soporte pig- mentadas y celulas sensoriales. En los restan- tes ojos (secundarios) se observaron tres tipos, sensoriales, celulas de soporte no pigmentadas y pigmentadas. En contraposicion a los dates conocidos para Salticidae y Lycosidae (Eakin & Bran- denburger 1971; Melamed & Trujillo-Cenoz 1966), las celulas sensoriales en la especie de Selenops estudiada se disponen en una sola capa en los OMA, en por lo menos dos capas en OLP y OMP y agrupadas latero-postero- extemamente en los OLA. En todos los ojos (excepto OMA) se puede observar mayor con- centracion de nucleos de las celulas sensori- ales en las zonas de convergencia de la luz, relacionado con la orientacion de cada ojo (Figs. 5, 9, 12). En los OMA las celulas de soporte pigmen- tadas son el unico tipo de elemento de soporte, actuando como celulas gliales y proporcion- ando una matriz de sosten de las celulas re- ceptoras y como una pantalla para la absor- cion de la luz. En los OLA, las celulas pigmentarias acompanan al nervio optico en una gran extension (Fig. 12). Los OLA, OLP y OMP ademas de las cel- ulas de soporte pigmentarias poseen las de so- porte no pigmentadas (Figs. 5, 9, 12), las que de acuerdo con Eakin & Brandenburger (1971) actuarian, ademas, como celulas nutri- cias y facilitarian el intercambio de sustancias entre las celulas gliales y los rabdomeros. Este tipo celular fue tambien observado por Me- lamed & Trujillo-Cenoz (1966) en Lycosidae. El tamafio de estas celulas, en Selenops, con respecto al tamano del ojo es muy destacable en los OLA (Fig. 12). Aparato dioptrico. — La forma de la su- perficie externa de la cornea estaria relacion- ada con la direccion de los rayos luminosos de acuerdo con la orientacion de los ojos; por lo que en los OMA, OLP y OMP la superficie externa es facetada (Fig. 11) mientras que la de los OLA es como placas imbricadas (Fig. 10). Las laminas observadas en la cornea y lente podrian, de acuerdo con Eakin & Bran- denburger (1971), actuar como un filtro de in- terferencia de diferentes longitudes de onda dentro de la luz blanca. Las diferentes formas de lentes observadas podrian influir sobre la refraccion de los rayos luminosos y su distri- bucion sobre la capa receptora del ojo. Las celulas cono, cuya funcion es la de se- cretar la lente (Eakin & Brandenburger 1971), en todos los ojos excepto en OLA, emiten pro- longaciones anteriores que cumplirian funcion del “humor vitreo” de los vertebrados. En los OLA, de acuerdo con Homann (1971) las cel- ulas cono se ubican inmediatamente por de- bajo de la lente, no existiendo proyecciones. La ausencia de “humor vitreo” en los OLA evitaria la absorcion de la luz en un medio mas denso, de tal manera que los rayos lu- minosos inciden sobre el tapete produciendo interferencia en pehcula delgada, redistribu- yendo la energia luminica y aumentando su intensidad hacia las celulas receptoras ubica- das en posicion latero-posterior-extema. De acuerdo con las observaciones de campo y la- boratorio esta especie es de habito crepuscular y noctumo. La funcion de los dos tipos dis- tintos de tapete observados en los ojos de esta especie se desconoce. La anatorma de los ojos secundarios ha sido utilizada para la sistematica de aranas y se sabe que las relaciones filogeneticas de Selen- opidae con otras familias son confusas. Leh- tinen (1967) ubica a los Cycloctenidae, Selen- opidae y Zoridae en su superfamilia Lycosoidea. Homann (1968) encuentra simi- litud en la estructura de los ojos de Cycloc- tenus con los de Selenops y posteriormente (Homann 1971) los ubica en Selenopidae y sugiere que estos taxa no estarian relacionados con los Lycosoidea. Levi (1982) considera por un lado a los Lycosoidea (Stiphidiidae, Acan- thoctenidae, Zoropsidae, Psechridae, Lycosi- dae, Pisauridae, Ctenidae, Senoculidae, Oxy- opidae y Toxopidae) y por otro a los Philodromoidea (Heteropodidae, Philodromi- dae, Selenopideae y Cycloctenidae). Griswold (1993) estudiando los Lycosoidea (todas las familias comprendidas en Levi (1982), excep- to Toxopidae) determina la monofilia del gru- po sobre la base de dos sinapomorfias, una de ellas es la presencia de tapete de tipo RT, en por lo menos uno de los ojos. Selenopidae y otros taxa previamente considerados como re- lacionados con uno o mas de los taxa tratados en ese estudio, no fueron incluidos porque de 48 THE JOURNAL OF ARACHNOLOGY acuerdo con Homann (1971) no estan rela- cionados con Lycosoidea. La presencia de tapete de tipo RT en los OLP y OMP de Selenops podria relacionarlos con los Lycosoidea {sensu Griswold 1993) y el hallazgo de un tapete KT en los OLA po- siblemente los relacionana con los Acanthoc= tenidae que tambien poseen este tipo de tapete en sus OLA. AGRADECIMIENTOS A la Lie. Marta Cardenas por sus comen- tarios sobre la interferencia luminica, a la Fundacion Miguel Lillo y al CONICET por el apoyo brindado. LITERATURA CITADA Eakin, R.M. & J.L. Brandenburger. 1971. Fine structure of the eyes of jumping spiders. J. Ul- trastruct. Res., 37:618-663. Griswold, C.E. 1993. Investigations into the phy- logeny of the lycosid spiders and their kin (Arachnida, Araneae, Lycosoidea). Smithson. Contrib. Zool., 539:1-39. Homann, H. 1968. Die stellung der Cyclocteninae im System der Araneen (Arachnida, Araneae). Senckerbergiana Biol. 49:51-57. Homann, H. 1971. Die augen der Araneae, Ana- tomic, Ontogenie und Bedeutung ftir die Syste- matik (Chelicerata, Arachnida). Z. Morph. Tiere, 69:201-272. Lehtinen, P.T. 1967. Classification of the cribellate spiders and some allied families with notes on the evolution of the Suborder Araneomorpha. Annal. Zool. Fennici, 4:199-468. Levi, H.W. 1982. Araneae. Pp. 2:77-95, In Syn- opsis and Classification of living organisms. (S.P. Parker, ed.). Melamed, J. «& O. Trujillo-Cenoz. 1966. The fine structure of the visual system of Lycosa (Ara- neae, Lycosidae). Z. fiir Zelfor., 74:12-31. Pickard-Cambridge, EO. 1900. Biologfa Centrali- Americana. Arachnida. Araneida. Opiliones. Vol. 11. Pp. 115-118. Cambridge. Simon, E. 1897. Histoire Naturelle des Araignees. Paris, Tome 2:27-61. Manuscript received 4 December 1995, revised 10 September 1996. 1997. The Journal of Arachnology 25:25:49-52 A NEW SPECIES OF CRYPSIDROMUS FROM BELIZE (ARANEAE, MYGALOMORPHAE, THERAPHOSIDAE) Steven B. RelcMing: Division of Ecology and Organismal Biology, The University of Memphis, Memphis, Tennessee 38152, USA ABSTRACT, A new species of Theraphosidae, Crypsidromus gutzkei, is described (from the male only) from northern Belize. Unique coloration in combination with pedipalps of intermediate length distinguishes C. gutzkei from all congeners. Character states proposed as diagnostic for males of Crypsidromus Ausserer 1871 and Metriopelma Becker 1878 are combined in the male of this new species, supporting the main- tenance of Metriopelma in the synonymy of Crypsidromus. The theraphosid genus Crypsidromus Aus- serer 1871 constitutes a taxon characterized by a dividing line of setae on tarsi IV and no tibial spurs on the mature male (Valerio 1980). Metriopelma Becker 1878 has often been treated as a synonym of Crypsidromus (Simon 1892; Petruekevitch 1911; Roewer 1942; Gerschman & Schiapeili 1973). Twelve spe- cies once considered as Metriopelma are thus included with nine other species under Cryp- sidromus. However, Valerio (1982) considered Metriopelma a valid taxon diagnosable from Crypsidromus by its fused, as opposed to dis- crete, spermathecae. Raven (1985) argued that the contentious apomorphy of the fused sper- mathecae rendered Crypsidromus without any autapomorpliic character, a situation he deemed untenable. Smith (1994) supported Valerio in restoring Metriopelma and suggest- ed that the two genera could be distinguished on the basis of spinatioe of the palpal tibia (fewer than four spines on the distal half in Metriopelma) and pedipalp length (longer in Metriopelma) in addition to spermathecal morphology. Sroith cautioned that validation of Metriopelma based on pedipalp character- istics would depend on the collection and ex- amination of additional specimens. The male of a new Crypsidromus species discovered in northern Belize exhibits a combination of Me- triopelma and Crypsidromus characters which supports Raven's synonymy. METHODS All measurements are in mm and were made using a dial caliper, ±0.01 mm. Leg and pedipalp measurements were taken from the left side. Trochanters and coxae were mea- sured from their ventral aspect while all other leg segment measurements were taken dorsal- ly. Description format follows Goloboff (1994). Spination abbreviations follow Pren- tice (1992). Standard abbreviations are used for ocular descriptions. Coloration was re- corded after specimen fixation under full spec- trum light using color charts in the Pantoee Book of Color (Eisman & Herbert 1990). Crypsidromus gutzkei new species Figs. 1-4, Table 1 Type.-=-Holotype male from Indian Church Village, Orange Walk District, Belize, 0. 1 km W of New River Lagoon, 1 October 1995, (S.B. Reichling). Holotype deposited in the American Museum of Natural History, New York. Etymology.—The specific epithet is a pat- ronym in honor of a superb biologist and the author’s scientific mentor, William H.N. Gutz- ke. Diagnosis. gutzkei new species is immediately discernible from most congeners by its uepattemed abdomen, as the genus is notable among New World thera- phosids for the number of species exhibiting bold abdominal patterns (Valerio 1980, 1982). The immaculate clothing of bright red setae on the abdomen of the holotype male is ge- nerically unique and distinguishes C gutzkei from all Central and South American conge- ners with unpattemed abdomens. The male of the Mexican C breyeri (Becker 1878), for 49 50 THE JOURNAL OF ARACHNOLOGY Table 1. — Leg measurements for the holotype male of Crypsidromus gutzkei new species. Mea- surements are in mm. Leg I II III IV Palp Coxa 6.3 4.6 3.8 4.4 3.7 Trochanter 2.2 2.0 1.8 1.7 1.8 Femur 11.3 11.3 10.0 11.6 8.3 Patella 3.8 5.4 4.9 5.0 4.5 Tibia 10.8 8.8 7.8 10.2 7.2 Metatarsus 7.8 7.9 9.0 14.0 Tarsus 5.9 5.7 5.6 6.3 1.7 Total 48.1 45.7 42.9 53.2 27.2 which the coloration in life is unknown, has longer pedipalps (exceeding tibia I in length, Smith 1994) than C. gutzkei. Crypsidromus gutzkei is further distin= guished from regional congeners by its com- paratively unmodified palpal embolus. In con- trast to other Central American Crypsidromus species, including species formerly assigned to Metriopelma (Valerio 1980, 1982), the api- cal division of the palpal embolus of C. gutz- kei is smoothly curved, as opposed to sharply bent, and the seminal groove lacks a promi- nent keel. The palpal embolus of C. gutzkei is most similar to C. brevibulbus Valerio 1980 from Costa Rica. However, C brevibulbus has a caput much wider than long (1.75X, Valerio 1980). Description. — Male (holotype): Length 27.7. Carapace length 12.8, width 10.3, cara- pace width/length 0.80; chelicerae, width 5.0; both fang furrows with twelve macroteeth; sternum, width 4.5, length 4.4; sigilla at base of coxae I, II, and III, posterior pair largest. Labial cuspules, 130; maxillary cuspules, 208, 200. Leg span, measured from apex of left tarsus I to apex of left tarsus IV, 106.5. Ped- ipalps extend to just beyond the basal third of tibiae 1. Leg and palp segment lengths in Ta- ble 1. Carapace clothed in iridescent pale gold (Pantone 15-0927) pubescence, closely ap- pressed. Dorsal surface of abdomen covered with long paprika-red (Pantone 17-1553) se- tae; short seal-brown (Pantone 19-1314) pu- bescence limited to patch of urticating hairs on posterior half of abdomen dorsum; ventral pubescence iron-gray (Pantone 18-1306). Dorsal and lateral surface of legs clothed in iridescent pale gold pubescence with scattered Figures 1, 2. — Crypsidromus gutzkei new spe- cies, male holotype. 1, Left palpal organ, retrola- teral view, showing apical keel (ak) bordering sem- inal groove; 2, Left palpal organ, prolateral view, showing gentle curve of embolus. Scale line = 1 mm. medium-length shale-gray (Pantone 19-3903) setae and sparsely scattered long beeswax-yel- low (Pantone 14-0941) setae which grade to seal-brown basally. Ventral surface of legs lighter, with gull-gray (Pantone 17-3802) pu- bescence and no golden setae. Fovea recurved. Anterior eye row slightly recurved; AME round, diameter 0.4, separated by 0.2; ALE ovoid, 0.3 X 0.4. Posterior eye row procurved; PME nearly round, diameter 0.2; PLE ovoid, 0.15 X 0.3, separated by 0.7. Caput length 1.6, width 1.2, length/width 1.33. Clypeus absent. Tibial spurs absent. All tarsi fully scopulate. Tarsi IV divided by a line of long, soft setae intermixed with dark, spi- niform setae. Extent of metatarsal scopulae: I, complete; II, 0.58; III, 0.24; IV, without scop- ulae. Palpal bulb length 2.4, width 1.2; simple, uniformly tapering embolus; apical division without prominent bend but with gentle down- ward curve; keel bordering seminal groove not prominent (Fig. 1); retrolateral surface of middle division concave with several angular changes in plane, prolateral surface convex and smooth; posterior face of basal division discretely inflated (Fig. 2). Spination: Leg I, metatarsus 2v(lam REICHLING— NEW CRYPSIDROMUS FROM BELIZE 51 Figures 3, 4. — CrypsMromus gutzkei new spe- cies, male holotype. 3, Left palpal tibia, dorsal view, showing two megaspines on the apical half; 4, Left palpal tibia, ventral view, showing three me- gaspines on the apical half. Scale line =1.5 mm. lmO.32), tibia 9v(2am lar lap lp036 lp0.30 lpO.24 2bp); leg II, metatarsus 6v(3am lpO.53 lrO.37 lpO.26), tibia llv(2am lar lap lr0.60 lmO.59 lpO.46 lm0.40 lr0.40 2bp); leg IH, metatarsus 4d(lam lap lmO.50 lpO.50) 12v(3am lar lap lep lrO.,61 lmO.46 lrO.33 2p033 Ibp), tibia 2d(lr0.45 Ibr) 5v(2am lap 2m0.50); leg IV, metatarsus 3d(lam lrO.50 lr0.40) 14v(3am lar lap 2em lpO.64 lr0.61 IrOJl lmO.49 lr036 lm0.21 ImO.ll), tibia 2d(lam lmO.15) 4v(2am lpO.69 lpO.29); palp, tibia 2d(lar ler) 5v(lam lap lrO.69 lpO.49 lp0.13). Distal half of palpal tibia with five megaspines (Figs. 3, 4). Female: Unknown. Distributioe.~~~Known only from the type locality. At present, C. gutzkei new species is the only CrypsMromus species reported from Belize. Relationships. -=“The spieatioe of the pal- pal tibia in combination with the intermediate length of the pedipalp apparent on the male of C gutzkei suggests that spermathecal mor- phology may be the only diagnostic character for separation of CrypsMromus and Me trio- peima. Since many CrypsMromus species and species formerly assigned to Metriopelma are known from very limited material, data de- scribing the range of variation occurring in spermathecal and palpal bulb morphology, and spination, is unavailable. Thus, unless fur- ther evidence is offered, Metriopelma should remain in the synonymy of CrypsMromus. Natural history.— The holotype was found roaming on a sloped river bank at 2230 h, during a light rain shower. The surrounding area was secondary tropical forest extensively fragmented by small agricultural plots. In light of the intensive research on theraphosid spider ecology that has been underway in the area for over a year, the fact that only one speci- men of this new taxon has been encountered suggests that C. gutzkei is rare. Material examined, — The holotype and the fol- lowing: Crypsidromus breyeri: MEXICO: Guana- juato (A, Duges) (BMNH). ACKNOWLEDGMENTS Financial support was provided by the Memphis Zoological Society Conservation Fund and by a Grant-in- Aid of Research from Sigma Xi, The Scientific Research Society. Field work and collection was conducted with permission from the Belize Ministry of Nat- ural Resources through the courtesy of E. Green, Chief Forest Officer. The manuscript was improved by the insightful comments of T Prentice and one anonymous reviewer. I am grateful to P. Hilly ard, British Museum of Nat- ural History (BMNFI) for the loan of speci- mens, and L. Leibensperger, Museum of Com- parative Zoology, Harvard University (MCZ) for supplying crucial references. Illustrations were the work of N. Reichlieg. I thank the Howells family for their generous hospitality. LITERATURE CITED Ausserer, A. 1871. Beitrage zur Kenntois der Ar- achniden-Familie der Territelariae Thorell (My- galidae Autor). Verhandl. K. K. ZooL-Bot. Ge- selL Wien, 21:177-224. Becker, L. 1878. Sur un noveau genre d'Avicular- iidae. Ann, Soc. Ent. Belgique, Compt. Rendus, 21:256-257. Eisman, L. & L. Herbert. 1990. The Pantone Book of Color. Hairy N. Abrams, Inc., New York. Gerschman, B..S. de Pikelin & R.D. Schiapelli. 1973. La subfamilia “Ischnocolinae” (Araneae: Theraphosidae). Rev. Mus. Argentino Cienc. Nat. Ent., 4:43-77. Goloboff, P.A. 1994. Linothele cavicola, a new di- plurine spider (Araneae, Dipluridae) from caves in Ecuador. J. ArachnoL, 22:70-72. Petrankevitch, A. 1911. A Synonymic Index-Cat- alogue of Spiders of North, Central and South America with all Adjacent Islands. Bull., Amer- ican Mus. Nat. Hist., 29:1-791. 52 THE JOURNAL OF ARACHNOLOGY Prentice, T.R. 1992. A new species of North Amer- ican tarantula, Aphonopeima paloma (Araneae, Mygalomorphae, Theraphosidae). J. ArachnoL, 20:189-199. Raven, R.J. 1985. The spider infraorder Mygalo- morphae (Araneae): cladistics and systematics. Bull. American Mus. Nat. Hist., 182:1-175. Roewer, C.E 1942. Katalog der Araneae. Bremen, vol. 1, 1040 pp. Simon, E. 1892. Etudes arachnologiques. 24 me- moire. XXXIX. Descriptions d’especes et de genres nouveaux de la famille des Aviculariidae (suite). Ann. Soc. Ent. France, 61:271-284. Smith, A.M. 1994. Theraphosid Spiders of the New World, vol. 2, Tarantulas of the USA and Mexico. Fitzgerald Publ., London. Valerio, C.E. 1980. Aranas terafosidas de Costa Rica (Araneae: Theraphosidae). III. Sphaero- bothria, Aphonopeima, Pterinopelma, Cithara- canthus, Crypsidromus y Stichoplastus. Rev. Biol. Trop., 28:271-296. Valerio, C.E. 1982. Aranas terafosidas de Costa Rica (Araneae, Theraphosidae). IV. Generos Me- triopelma y Cyclosternum, incluyendo especies de Panama. Brenesia, 19/20:407-423. Manuscript received 14 November 1995, revised 15 June 1996. 1997. The Journal of Arachnology 25:53-83 THE SPIDER FAMILY CYATHOLIPIDAE IN MADAGASCAR (ARANEAE, ARANEOIDEA) Charles E. Griswold: Department of Entomology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118 USA ABSTRACT. The family Cyatholipidae is newly recorded from Madagascar, including the following new taxa: Ulwembua ranomafana new species, and Ulwembua antsiranana new species; Vazaha toa- masina new genus new species; and Alaranea new genus, including Alaranea betsileo, Alaranea alba, Alaranea merina, and Alaranea ardua, all new species. Madagascar is widely recognized as being of great conservation importance (National Research Council 1980; Rasoanaivo 1990) be- cause the island is known for high rates of endemism and unique occurrence of primitive members of otherwise widespread taxa (My- ers 1988). Ongoing rapid habitat destruction, particularly of forests, makes the collection, description, and study of the evolutionary and biogeographic significance of the Malagasy biota particularly urgent. Nevertheless, the spider fauna of Madagascar remains poorly known. The number of spider species record- ed from the whole island only slightly exceeds 400 (V. Roth in lit.), significantly less than the 626 species recorded from the British Isles (Merrett, Locket & Millidge 1985; Merrett & Millidge 1992). Yet, nearly 400 species have been collected from a single site in the south- ern part of the island (V. Roth pers. comm.), suggesting a rich fauna. Alderweireldt & Jocque (1994) suggest that the known com- ponent of the Malagasy spiders fauna is around 10%, a figure rendered more credible by the recent discovery of a hitherto unknown spider family (Jocque 1994). Given the cur- rent state of our knowledge, the discovery of Cyatholipidae in Madagascar is not surprising. The Cyatholipidae were previously known from Africa, Jamaica, New Zealand and Aus- tralia (Griswold 1987; Forster 1988; Simon 1894). Three species from Baltic amber are attributed to this family (Wunderlich 1993). Seven new species belonging to three genera, two of them new, are herein described from Madagascar. Two new species, Ulwembua ranomafana and U. antsiranana, have their congeners in southern Africa. A new genus, Alaranea, appears to be closely related to an undescribed genus from the mountains of eastern Africa. The affinities of the new genus Vazaha are enigmatic. All of the new species occur in moist for- ests, where some may be very common. The new species Alaranea betsileo, A. merina, and Ulwembua antsiranana are among the most common arboreal spiders in the forests where they occur. Dozens may be found in an hour of collecting. All hang beneath sheet webs (see Davies 1978). That these common spiders were previously undescribed underscores the poor state of our current knowledge of the Madagascar spider fauna. The material upon which this study was based was largely collected by the author and colleagues Nikolaj Scharff, Jonathan Cod- dington, Scott Larcher and Rija Andriamasa- manana during October-December 1993. Most material collected during that period is divided among the California Academy of Sciences (CAS), Zoological Museum, Univer- sity of Copenhagen (ZMUC), and Smithsoni- an Institution, Washington D.C. (USNM). Ad- ditional material was made available through the courtesy of J. Coddington of the USNM, R. Jocque of the Musee Royal de UAfrique Centrale, Tervuren (MRAC), H. Levi of the Museum of Comparative Zoology, Harvard (MCZ), C. Rollard of the Museum National d’Histoire Naturelle, Paris (MNHN), and Vin- cent and Barbara Roth. METHODS Prior to examination with a Hitachi S-520 scanning electron microscope aU stmctures were 53 54 THE JOURNAL OF ARACHNOLOGY Figure 1. — Ulwembua antsiranana new species, holotype male, lateral view, (Scale bar = 1 mm) critical point dried. Vulvae were cleaned by ex- posure to trypsin, bleached in “CMorox” house- hold bleach (5.25% sodium hypochlorite), stained with Chlorazol Black, and mounted in Hoyer's Medium for examination and photog- raphy. Examination was via Wild M5Apo and Leitz Ortholux II microscopes, and photography of vulvae was by an Olympus PM- 1 OAK at- tached to the Leitz Ortholux II. Small structures were examined in temporary mounts as de- scribed in Coddington (1983). Abbreviations are listed in Table 1. All measurements are in mm. Specimens mea- sured were chosen to encompass largest and smallest individuals. TAXONOMY Cyatholipidae Simon 1894 Cyatholipeae Simon 1894:711, based on Cyatholi- pus hirsutissimus Simon 1894. Roewer 1942:967. Cyatholipinae, Wunderlich 1978:33. Teemenaaridae Davies 1978:42, based on Teeme- naarus silvestris Davies 1978. Cyatholipidae, Platnick 1979:116. Brignoli 1983:231. Griswold 1987:501. Forster 1988:7. Platnick 1989: 181. Platnick 1993:172. Wuncterlich 1993:234. Diagnosis.— Colulate, entelegyne araneoids that share with the Synotaxidae a cup-shaped paracymbium (Fig. 33) and posteriorly broad- ly truncate sternum (Figs. 49, 67), and differ- ing in having a retromedian cymbial process (Figs. 17, 33) and very broad posterior respi- ratory groove (Fig. 52). Description.— For full description see Griswold (1987) and Forster (1988). Total length 1-4 mm; labium broader than long (Fig. 49); chelicerae smooth laterally with three small retromarginal and, in most taxa, four large promarginal teeth (Fig. 6); legs spineless (Figs. 1, 15, 68), ITC short (Fig. 3); tarsal organ (Fig. 8) and trichobothrial bases (Fig. 7) round and smooth; spinning organs (Figs. 9-14) typical of the Araneoidea in hav- ing a single ALS major ampullate gland spigot plus nubbin and 12-14 piriform gland spigots with highly reduced bases; PMS with large, anteromedian cylindrical gland spigot, two aciniform gland spigots, and posterior minor ampullate gland spigot, CY spigot absent in male; PLS with araneoid triplet of one flagel- liform gland and two aggregate gland spigots, two AC spigots, and a single mesal CY spigot GRISWOLD— MADAGASCAR CYATHOLIPIDAE 55 Figures 2-8. — Morphology of Alaranea spp. 2, 4. Carapace and abdomen, lateral view; 3. Apex of tarsus I, showing claws; 5. Epiandrous region: epigastric furrow with epiandrous spigots (upper), close-up of epiandrous spigots (lower); 6. Cheliceral fang furrow, anterior; 7. Trichobothrial base, tibia I; 8. Tarsal organ, palpus; 2, 8. Alaranea betsileo new species, male from Talatakely; 3, 4, 6, 7. Alaranea betsileo new species, female from Talatakely; 5. Alaranea merina new species, male from Perinet. 56 THE JOURNAL OF ARACHNOLOGY KEY TO THE CYATHOLIPIDAE OF MADAGASCAR 1. 2(1). 3(2). 4(1). 5(4). 6(5). 7(6). 8(4). 9(8). Abdomen not sclerotized around base of pedicel, male lacking scutum; parembolic process absent; coxae separated by soft cuticle, pleural and sternal sclerotizations separate (Figs. 1, 15, 41) ..................................................................... 2 Abdomen sclerotized completely around base of pedicel to form annulate petiole produced dorsally into a short projection or horn (Figs. 4, 68, 94); abdomen of males with a thin, shiny transparent dorsal scutum (Fig. 95); parembolic process present (Figs. 60, 62, 73); pleural and sternal sclerotizations meet to surround coxae (Figs. 68, 94) (Alaranea new genus) ......... 4 Chelicerae with basal projection small or lacking; epigynum with median hood (Figs. 18, 19); apex of cymbial RMP directed ventrad, well separated from PC (Figs. 22, 33) (Ulwembua) ... 3 Chelicerae with large basal projection (Fig. 41); epigynum lacking median hood (Figs. 29, 30, 43); apex of cymbial RMP directed distad, juxtaposed to PC (Figs. 44, 48) ............. ......................................... Vazaha toamasima new genus, new species Conductor simple (Figs. 20, 34); carapace dark except along lateral margins and on central longitudinal band extending from posterior median eyes posteriorly to behind thoracic fovea (Fig. 39); afferant duct of vulva with three loops (Fig. 36) Ulwembua ranomafana new species Conductor bipartite (Figs. 16, '23); carapace light except ocular area, margins of pars cephalica, and diffuse radii from thoracic fovea on pars thoracica dark (Fig. 38); afferant duct of vulva with five loops (Fig. 35) ........................... Ulwembua antsiranana new species Males ................................................................... 5 Females ................................................................. 8 Conductor simple (Figs. 72, 79, 88) ............................................ 6 Conductor bipartite, with thin, broad proximal piece separate from conductor proper (Fig. 61) Alaranea betsileo new species Proximal point of conductor no longer than distal cup (Figs. 72, 88) .................... 7 Proximal point of conductor elongate attenuate (Figs. 58, 79) ...... Alaranea alba new species Proximal point of conductor small, narrower than cup (Fig. 72) ... Alaranea merino new species Proximal point of conductor thick, bifid, equal in width to cup (Fig. 88) ............... .................................................... Alaranea ardua new species Sternum dark red-brown to black (Fig. 97), abdomen of most specimens with extensive dark markings ................................................................. 9 Sternum pale yellow-brown, abdomen white, marked with lateral, ventral, and posterior black spots (Figs. 67-69) ...................................... Alaranea alba new species Dorsum of abdomen (Figs. 65, 66, 95, 96) with longitudinal dark bands diverging from apex to middle and converging posteriorly (these bands may be bold, faint, or almost completely obscured by dark markings) .................. Alaranea merina and A. ardua new species Dorsum of abdomen (Figs. 63, 64) lacking such marks, most specimens with median black band surrounding 1-2 anterior white spots .................. Alaranea betsileo new species (basal CY spigot universally absent in fe- males); males retain triplet; colulus a trian- gular, fleshy lobe (Fig. 52); male epiandrous spigots scattered in groups of two to four an- terior of epigastric furrow (Fig. 5); cymbium of male palpal tarsus with basal, cup-shaped paracymbium and retromedian process along the retrolateral margin of the cymbium just distad of the PC (Figs. 31, 33, 48, 70); palpal bulb (Figs. 31-34) with flattened, cup-shaped subtegulum and round to oval, convex tegul- um; T with apical lobe, in most taxa produced ventromediaily into blunt to pointed, dentate tegular lobe; T with median conductor, simple or consisting of two processes (e.g., in Ul- wembua antsiranana. Fig. 23, and Alaranea betsileo, Fig. 61); embolus spirals clockwise (left palp, ventral view), making nearly full turn, thick with tmncus and pars pendula clearly distinguished; may or may not be a parembolic process at % the length of the E (Figs. 60, 62, 73); epigynum (Figs. 25-30) of most taxa with anterior, projecting scape, pos- teriad of this a depressed atrium with trans- verse, median hood hiding copulatory open- ings that are separated by an interior median septum; cuticle laterad of epigynum probably homologous to lateral lobes of other epigyna, these may form narrow, inward-curving pro- cesses along epigastric furrow that disappear anteriorly beneath the MH; the area between these processes comprises the epigynal me- dian lobe; vulva (Figs. 35-37) with postero- ventral copulatory openings opening into an- GRISWOLD—MADAGASCAR CYATHOLIPIDAE 57 Table 1. — List of anatomical abbreviations used in the text and figures. AC acieiform gland spigot(s) AD vulval afferant duct AER anterior eye row AG aggregate gland spigot(s) AL anterior lateral eyes ALS anterior lateral spinneret A apical lobe of tegulum AM anterior median eyes AT epigyeal atrium C conductor CB cymbium CO copulatory opening CY cylindrical gland spigot(s) E embolus EF epigastric furrow FD fertilization duct FL flagelliform gland spigot(s) HS spermathecal head ITC inferior tarsal claw LL epigynal lateral lobes MAP major ampullate gland spigot(s) mAP minor ampullate gland spigot(s) MH epigynal median hood ML epigynal median lobe MS epigynal median septum OAL ocular area length OQA ocular quadrangle, anterior OQP ocular, quadrangle, posterior PC paracymbium PER posterior eye row PI piriform gland spigot(s) PLS posterior lateral spinneret PL posterior lateral eyes PM posterior median eyes PMS posterior median spinneret PP parembolic process RMP retromediae cymbial process S epigynal scape ST subtegulum T tegulum TL ventromedian tegular lobe teriad“directed afferent duct, AD sclerotized (Figs 90-93) or hyaline (Figs. 35, 36), simple or elaborately folded, or rarely absent (Fig, 37); spermathecal head of most taxa dorsad of CO and entered laterally by AD, heavily scler- otized, nearly spherical with anterior pores; fertilization duct posterior. Uiwembua Griswold 1987 Ulwembua Griswold 1987:532. Type species, by original designation, Ulwembua puichra Gris- wold 1987. Platnick 1989:182. Diagnosis,— Abdomen triangular (Figs. 1, 15); coxae not surrounded by sclerotization; legs long, length femur I greater than 2.5 X carapace width; carapace with dorsal light mark (Fig, 39); palpus lacking PP (Figs, 16, 20); vulva with extensive, hyaline AD (Figs. 35, 36). Description (encompasses all members of genus).— Total length 2.00-3.32. Carapace oval in dorsal view (Figs. 38, 39), length 1.39-1.61 times width, low in most species, maximum height 0.41-0.51 width; texture finely rugose to granulate, in most specimens becoming denticulate posteriorly, thoracic fo- vea oval to round, indistinct, shallow in fe- male and deeper in male; carapace posterior margin truncate to weakly concave; ocular area with PER width 1.95-2.50 times OAL, 2.30-2,80 times OQP, OQP 0.83-1.07 times OQA; diameter AM 1.00-1.80 times PM, dis- tance PM-PL 1.07-1.85 times PM diameter; clypeal height 1.22-3.21- times AM diameter, cheliceral length 1.84-3.20 times clypeal height; chelicerae unmodified or with basal projection {Ulwembua ranomafana). Sternum rugose to pustulate, length 0.96-1.14 times width, coxae surrounded by unsclerotized cu- ticle (Figs. 1, 15). Abdomen triangular, un- sclerotized or sclerotized around pedicel, not petiolate; abdominal setae short, slender, bases of anterior setae slightly enlarged. Legs long, femur I 2. 5-4. 5 times carapace width, unmod- ified. Male palpus (Figs. 16, 17, 20-24, 31- 34) with cymbial RMP pointing ventrad, smaller than PC; palpal bulb with dentate TL, apex a small, smooth to pustulate lobe; C smooth, variable, median or basal, longitudi- nal or transverse, simple or with accessory process; E thick, long, in most species em- bolus makes more than 1.1 rotation, base smooth, simple, origin apical near 12 o'clock; PP absent; spermduct with tight double twist (curlicue) near embolic base. Epigyeum (Figs. 18, 19, 25-28) with S and MH, septum be- tween copulatory openings slender to broad, atrial furrows may or may not extend behind S; ML parallel-sided. Vulva (Figs. 35, 36) with extensive hyaline AD, extending anteriad then posteriad to join HS; FD posterior. Composition.- — ^Five species, two in Mad- agascar. Distribution.- — Southern Africa; Madagas- car (Fig. 98). 58 THE JOURNAL OF ARACHNOLOGY Figures 9-14 — Spinnerets of Alaramea betsileo new species, from Talatakely. 9, 10. ALS; 11, 12. PMS; 13, 14. PLS; 9, 11, 13. Female; 10, 12, 14, Male. AC = aciniform gland spigots; AG = aggregate gland spigots; CY = cylindrical gland spigots; FL = flagelliform gland spigot; MAP - major ampullate gland spigot; mAP = minor ampullate gland spigot; PI = piriform gland spigots. GRISWOLD— MADAGASCAR CYATHOLIPIDAE 59 Figure 15. — Ulwembua ranomafana new species, paratype female, lateral view. (Scale bar = 1 mm) Ulwembua antsiranana new species Figs. 1, 16-=18, 22-26, 35, 38, 98 Types. — Male holotype and female para- type from forest at an elevation of approxi- mately 1000 m at Parc National Montagne d’Ambre (12°32'S, 49°10'E), Antsiranana Province, Madagascar, 30 November 1993, C.E. Griswold (CAS). Etymology. — -Antsiranana, the province of the type locality, a noun in apposition to the generic name. Diagnosis.— Carapace light except dark on ocular area, margins of pars cephalica, and diffuse radii from thoracic fovea on pars thor- acica (Fig. 38). Male with E strongly sinuate across tegulum base, C double (Figs. 16, 23). Vulva with AD complex, having five loops (Fig. 35). Description.— Mafe (holotype): As in Fig. 1. Total length 2.66. Carapace dusky yellow- gray along lateral margin, gray narrowing be- hind ocular area, and along margins of pars cephalica, faintly mottled in center, with dark gray forming narrow longitudinal band anter- iad of thoracic fovea and faint bands radiating from thoracic fovea to margin, dorsum be- tween these marks yellow-brown; ocular area with black surrounding and extending be- tween AM and posteriad to surround each PM, and surrounding and extending between lateral eyes; clypeus yellow-brown, dark in center from AM to oral margin; chelicerae and palpal coxae brown, labium and sternum near- ly black, unmarked; coxae, trochanters, legs, and palpi yellow-white, cymbium dark brown, legs shading to yellow-gray from distally on femora to tarsi, unmarked; abdomen white, dorsum with pair of median and lateral dark gray longitudinal bands that meet at abdomi- nal apex, venter gray from abdominal apex to pedicel. Carapace 1.08 long, 0.67 wide, 0.35 high, texture finely granulate, posterior margin weakly concave; thoracic fovea round, very shallow, with small posterior pit; PER and 60 THE JOURNAL OF ARACHNOLOGY Figures 16-21, — Genitalia of Ulwembua spp. 16, 17, 20, 21. Left male palpus; 18, 19. Epigynum; 16, 18-20. Ventral view; 17, 21. Retrolateral view; 16, 17. Ulwembua antsiranana new species, holotype; 18. £/. antsiranana new species, Moetagne d’Ambre; 19. £/. ranomafana new species, paratype; 20, 21. U. ranomafana new species, holotype. GRISWOLD— MADAGASCAR CYATHOLIPIDAE 61 Figures 22-24. — Right male palpus of Ulwembua antsiranana new species, Montagne d'Ambre. 22. Retrolateral view; 23. Ventral view; 24. Prolateral view. AER 0.42 wide, OAL 0.20; ratio AM:AL:PM: PL, 1.33:1.08:1.0:1.17, PM diameter 0.06. Clypeus 0.18 high, chelicerae 0.35 long, un- marked. Sternum 0.58 long, 0.56 wide, ru- gose; labium 0.1 1 long, 0.19 wide; palpal cox- ae 0.20 long, 0.13 wide. Leg measurements (femur + patella + tibia + metatarsus + tar- sus - [Total]): I: 2.85 + 0.30 + 2.62 + 2.42 + 1.19 = [9.38]; II: 2.13 + 0.25 + 1.76 + 1.55 + 0.87 - [6.56]; III: 1.13 + 0.23 + 0.85 + 0.83 + 0.53 = [3.57]; IV: 1.70 + 0.25 + 1.32 + 1.08 + 0.62 = [4.97]; Palp: 0.29 + 0.11 + 0.10 + (absent) + 0.38 = [0.88]. Ab- domen unsclerotized except between epigas- tric furrow and pedicle. Palp (Figs. 16, 17, 22-™24) with cymbial RMP short, narrow, pointed, PC a narrow hook in lateral view; tegulum apex low, smooth, TL small, convex, with small oval denticulate patch; C large, smooth, with small, narrow basal article. Variation: (n = 3) Total length 2.18-2.66; ratios of carapace length/width 1.50-1.61, height/width 0.48-0.51; ratios of PER/OQP 2.56-2.69, PER/OAL 2.05-2.10, OQP/OQA 0.83-1.00, PM“PL distance/PM diameter 1.50-1.82, diameter AM/PM 1.17-1.33; ratios of clypeal height/diameter AM 2.12-2.86, cheliceral length/clypeal height 1.84-2.05; ra- tio of sternum length/width 1.04-1.09; ratio of femur I length/carapace width 4.00-4.65. Dorsal longitudinal bands of abdomen narrow to broad, separate from to confluent with lat- eral longitudinal bands. Female (paratype): Total length 2.85. Mark- ings and structure as in male (Fig. 38). Carapace 1.05 long, 0.68 wide, 0.30 high; PER and AER 0.44 wide, OAL 0.20; ratio AM:AL:PM:PL, 1.6:1.0:1.0:1.08, PM diameter 0.05. Clypeus 0.13 high, chehcerae 0.39 long. Sternum 0.61 long, 0.55 wide; labium 0.11 long, 0.20 wide; palpal coxae 0.21 long, 0.17 wide. Leg mea- surements (femur -H patella + tibia H- metatar- sus + tarsus = [Total]): 1: 2.72 + 0.30 + 2.45 + 2.21 + 1.08 - [8.76]; H: 2.00 + 0.25 -f 1.57 + 1.40 + 0.85 = [6.07]; HI: 1.02 + 0.23 + 0.74 + 0.74 + 0.53 = [3.26]; IV: 1.59 + 0.28 + 1.23 + 1.02 F 0.57 = [4.69]; Palp: 0.23 + 0.10 + 0.14 + (absent) + 0.34 — [0.81]. Epi- gynum as in Figs. 18, 25, 26, MS slender, atrial fiirrows end at S; vulva as in Fig. 35, hyahne AD having small anteromedian fold, large an- terior fold, and three small lateral folds before joining HS, Variation: {n = 3). Total length 2.25-3.28; ratios of carapace length/width 1.49-1.54, height/width 0.45-0.49; ratios of PER/OQP 2.60-2.80, PER/OAL 2.11-2.33, OQP/OQA 0.88-0.94, PM-PL distance/PM diameter 1.64-1.80, diameter AM/PM 1.60-1.80; ratios of clypeal height/diameter AM 1.22- 1.55, cheliceral length/clypeal height 3.08- 3.18; ratio of sternum length/width 1.04-1.11; ratio of femur I length/carapace width 3.36- 3.94. Dorsal longitudinal bands of abdomen narrow to broad, separate from to confluent with lateral longitudinal bands. Natural History. =These spiders were common in wet montane forest. Individuals built sheet webs in low vegetation, rarely more than 30-40 cm from the forest floor. Distribution.- — Known only from the type locality, an isolated montane rain forest in northern Madagascar (Fig. 98). 62 THE JOURNAL OF ARACHNOLOGY Figures 25-30. — Epigyna of Cyatholipidae. 25, 27, 29. Ventral view; 26, 28, 30. Lateral view; 25, 26. Ulwembua antsiranana new species, Montagne d’Ambre; 27, 28. Ulwembua ranomafana new species, paratype; 29, 30. Vazaha toamasina new species, paratype. AT = atrium; CO = copulatory openings; EF = epigastric furrow; LL = lateral lobes; MH = median hood; ML = median lobe; MS = median septum; S = scape. GRISWOLD— MADAGASCAR CYATHOLIPIDAE 63 Figures 31-34. — Right male palpus of Ulwembua ranomafana new species, holotype. 31. Retrolateral view; 32. Proventral view; 33. Cymbial base, retrolateral view; 34. Ventral view. A = apical lobe of tegulum; C = conductor; CB = cymbium; E = embolus; PC = paracymbium; RMP = retromedian cymbial process; ST = subtegulum; T = tegulum; TL = ventromedian tegular lobe. Additional material examined. — MADAGAS- CAR: Antsiranana Province: Parc National Mon- tagne d’Ambre, 2.79 air km NE of park entrance, forest, (12°32'S, 49°10'E), elev. approx. 1000 m, 21-30 November 1993 (N. Scharff, C. Griswold, J. Coddington, S. Larcher and R. Andriamasaman- ana), 31(367$, one pair in MRAC, remainder in CAS, USNM, and ZMUC. Ulwembua ranomafana new species Figs. 15, 19-21, 27, 28, 31-34, 36, 39, 98 Types.^ — ^Male holotype and female para- type from forest at approximately 1100 m el- evation at Vohiparara, Parc National de Ran- omafana, Fianarantsoa Province, Madagascar, 7 December 1993, C. Griswold (CAS). Etymology.— The type locality, a noun in apposition to the generic name. Diagnosis. — Carapace dark except along lateral margins and on central longitudinal band extending from PM posteriorly to behind thoracic fovea (Fig. 39). Male with E weakly sinuate across tegulum base, C simple, medi- an, longitudinal (Figs. 20, 34). Vulva with AD simpler than in U. antsiranana (Fig. 36). 64 THE JOURNAL OF ARACHNOLOGY Figures 35-37. — Vulvae of Cyatholipidae, dorsal view, cleared. 35. Ulwembua antsiranana new spe- cies, Montagne d’Ambre; 36. U. ranomafana new species, paratype; 37. Vazaha toamasina new species, paratype, AD = afferent duct; FD = fertilization duct; HS = spermathecal head. Description. — Male (holotype): Total length 2.47. Carapace yellow-white, with broad dark gray dorsolateral bands extending from margins of pars cephalica to posterior margin, leaving narrow yellow-white band along lateral margin and broad central yellow- white band from pars cephalica to thoracic fo- vea; ocular area with black surrounding and extending between AM and extending poste- riad to surround each PM, and surrounding and extending between lateral eyes; clypeus yellow-brown, dark in center from AM to oral margin; chelicerae dark brown, palpal coxae, labium and sternum nearly black; coxae, tro- chanters, basal segments of palpi and bases of leg femora yellow-white, cymbium dark brown, legs shading distally to yellow-brown, unmarked except that apices of femora and tibiae are lighter; abdomen white, with black dorsolateral bands meeting posteriorly, each band encompasses narrow anterior and medi- an oval white spots, venter gray, black from spinnerets to pedicel. Carapace 1.24 long, 0.80 wide, 0.36 high, texture finely granulate becoming denticulate posteriorly, posterior margin truncate, thoracic fovea a deep oval; PER 0.47 wide, AER 0.46 wide, OAL 0.22; ratio AM:AL:PM:PL, 1.23:1.08:1.0:1.23, PM diameter 0.07. Clypeus 0.18 high, chelicerae 0.37 long, with basal projection. Sternum 0.61 long, 0.55 wide, rugose; labium 0.11 long, 0.20 wide; palpal coxae 0,21 long, 0.17 wide. Leg measurements (femur + patella + tibia + metatarsus + tarsus ~ [Total]): I: 3.74 + 0.34 + 3.51 + 3.72 + 1.49 = [12.80]; II: 2.47 + 0.30 + 1.94 + 2.04 + 0.96 = [7.71]; III: 0.96 + 0.21 + 0.79 + 0.79 + 0.49 = [3.24]; IV: 1.59 + 0.25 + 1.23 + 1.06 + 0.53 = [4.66]; Palp: 0.35 + 0.13 + 0.10 + (absent) + 0.40 = [0.98]. Abdomen unsclerotized except strongly between epigastric furrow and pedi- cle. Palp (Figs. 20, 21, 31-34) with cymbial RMP blunt, very short, PC broad in lateral view; tegulum apex bulging, smooth, TL small, denticulate in elongate oval patch; C simple. Female (paratype): As in Figs. 15, 39. Total length 2.85. Markings and structure as in male except dorsal light marking of carapace broad- er, black dorsolateral bands of abdomen en- compassing broad lateral white spots, anterior white spots confluent with median white band, venter gray. Carapace 1.05 long, 0.68 wide, 0.30 high, thoracic fovea a shallow oval; PER and AER 0.44 wide, OAL 0.20; ratio AM:AL: PM:PL, 1.6: 1.2: 1.0: 1.3, PM diameter 0.05. Clypeus 0.13 high, chelicerae 0.39 long, with weak basal projection. Sternum 0.61 long, 0.55 wide; labium 0.11 long, 0,20 wide; palpal coxae 0.21 long, 0.17 wide. Leg measure- ments (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 2.51 + 0.28 + 2,13 + 2.13 A 1.04 = [8.09]; II: 1.55 + 0.23 + 1.13 + 1.21 + 0.70 = [4.82]; III: 0.79 + 0.17 -h 0.51 + 0.45 + 0.42 = [2.34]; IV: 1.28 + 0.23 + 0.91 + 0.79 + 0.49 = [2.34]; Palp: 0.24 + 0.10 + 0.13 + (absent) + 0.33 = [0.80]. Epi- gynum as in Figs. 19, 27, 28; MS between CO broad, atrial furrows end just behind S; vulva as in Fig. 36, hyaline AD having broad anteromedian chamber and forming large lat- eral and posterolateral folds before joining HS. Variation: (n — 3). Total length 2.72-3.19; GRISWOLD— MADAGASCAR CYATHOLIPIDAE 65 Figures 38-40. — Female Cyatholipidae, dorsal views. 38. Ulwembua antsiranana new species, Mon- tague d’Ambre; 39. U. ranomafana new species, paratype; 40. Vazaha toamasina new species, paratype. (Scale bar = 1 mm) ratios of carapace length/width 1.39-1.50, height/width 0.38-0.50; ratios of PER/OQP 2.35-2.62, PER/OAL 1.95-2.21, OQP/OQA 0.89-1.00, PM“PL distance/PM diameter 1.07-1.33, diameter AM/PM 1.14-1.50; ratios of clypeal height/diameter AM 1.44-1.50, cheliceral length/clypeal height 2.61-3.00; ra- tio of sternum length/width 1.00-1.14; ratio of length femur I/carapace width 3.57-3.71. Distribution. — Known only from the type locality in montane rain forest (Fig. 98). Additional material examined. — MADAGAS- CAR: Fianarantsoa Province: Parc National de Ranomafana, Vohiparara, ca. 21°14'S, 47°24'E, elev. 1100 m, 5-7 November 1993 (N. Scharff, S. Larcher, C. Griswold, and R. Andriamasamanana) 29 (ZMUC, USNM). Vazaha new genus Type species. — -Vazaha toamasina new species Etymology. — -From the Malagasy for for- eigner; gender feminine. Diagnosis. — ^Female epigynum (Figs. 29, 30, 43) with S but lacking MH, male palp with cymbial RMP directed distad (Figs. 45, 48); E thick, lacking PP. Description.— See under species descrip- tion below of Vazaha toamasina new species. Composition. — One species. Distribution. — Madagascar (Fig. 98). Vazaha toamasina new species Figs. 29, 30, 37, 40-48, 98 Types. — Male holotype and female para- type from forest at an elevation of 1000 m at Parc National Perinet, Toamasina Province, Madagascar, 4-5 November 1993, C. E. Gris- wold (CAS). Etymology. — From the home province, a noun in apposition to the generic name. Diagnosis.— -See generic diagnosis above. Description. — Male (holotype): As in Fig- ure 41. Total length 2.38. Carapace dusky yel- low-brown, pale yellow-brown in center be- hind pars cephalica surrounding thoracic fovea, ocular area dusky, dark grey surround- ing AME; clypeus and chelicerae dusky grey- brown, unmarked, palpal coxae, labium and sternum dark grey-brown; coxae and trochan- ters white, legs shading from white at base to pale yellow-brown from distal femora to tarsi, palpi white except for grey-brown cymbium; abdomen pale grey, dorsum with median pair of longitudinal dark bands that meet posteri- orly outlining median white area and laterally outlining an anterior and median light longi- tudinal spot, venter dark grey from pedicel to beyond spinnerets, weakly sclerotized ven- trally between pedicel and epigastric furrow. Figure 41. — Vazaha toamasina new species, holotype male, lateral view. (Scale bar = 1 mm) otherwise unsclerotized, abdominal setae fine, with small setal base picks on anterior margin. Carapace 1.09 long, 0.68 wide, 0.35 high, oval in dorsal view, posterior margin weakly con- cave, finely rugose, thoracic fovea a shallow oval; PER and AER 0.43 wide, OAL 0.20; ratio AM:AL:PM:PL, 1.0:1.33:1.0:0.92, PM diameter 0.06. Clypeus 0.16 high, chelicerae 0.46 long, with large anteriad-directed basal projection. Sternum 0.60 long, 0.55 wide, finely granulate; labium 0.11 long, 0.18 wide; palpal coxae 0.21 long, 0.16 wide; leg coxae surrounded by unsclerotized cuticle. Leg mea- surements (femur + patella + tibia -f meta- tarsus + tarsus = [Total]): I: 2.64 + 0.23 + 2.38 + 2.36 + (missing) - [??]; II: 1.94 + 0.25 + 1.59 + 1.57 + 0.81 = [6.16]; III: 0.89 + 0.21 + 0.68 + 0.68 + 0.47 - [2.93]; IV: 1.42 + 0.23 + 1.00 + 1.00 + 0.49 - [4.14]; Palp: 0.33 + 0.13 + 0.08 + (absent) + 0.26 = [0.80]. Palp (Figs. 42, 44-48) with cymbial RMP short, narrow, directed distad, PC broad and blunt in lateral view; tegulum apex a small, pointed lobe, TL pointed, weakly wrin- kled; C a distal, simple, elongate basad-di- rected triangle; E stout, arising near 2 o’clock, PP absent. Female (paratype): As in Fig. 40. Total length 2.21. Markings as in male. Carapace 1.01 long, 0.70 wide, 0.31 high; PER 0.40 wide, AER 0.38 wide, OAL 0.19; ratio AM: AL:PM:PL, 1.17:1.0:1.0:1.0, PM diameter 0.06. Clypeus 0.11 high, chelicerae 0.39 long, with small anteriad-directed basal projection. Sternum 0.59 long, 0.51 wide, texture nearly smooth; labium 0.14 long, 0.18 wide; palpal coxae 0.21 long, 0.15 wide. Leg measure- ments (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 2.29 + 0.25 + 1.91 + 1.91 + 0.91 = [7.27]; II: 1.62 + 0.28 + 1.28 + 1.28 + 0.72 = [5.18]; III: 0.74 + 0.21 + 0.57 + 0.57 + 0.42 = [2.51]; IV: 1.28 + 0.21 + 0.96 + 0.81 + 0.47 - [3.73]; Palp: 0.25 T 0.08 + 0.11 + (absent) + 0.28 = [0.72]. Epi- gynum as in Figs. 29, 30, 43, with S but lack- ing MH; vulva as in Fig. 37, CO lead directly to large, sclerotized HS, AD absent. Variation: (n = 2). Total length 2.21-2.81; GRISWOLD— MADAGASCAR CYATHOLIPIDAE 67 Figures 42-44. — Genitalia of Vazaha toamasina new species. 42, 44. Left male palpus, holotype; 43. Epigynum, paratype; 42, 43. Ventral view; 44. Retrolateral view. ratios of carapace length/width L43-L51, height/width 0.43-0.45; ratios of PER/OQP 2.53-2.78, PER/OAL 2.11-2.17, OQP/OQA 0.87-1.00, diameter AM/PM 1.17-1.45; ratios of clypeal height/diameter AM 1.25-1.57, cheliceral length/clypeal height 3.36-4.00; ra- tio of length femur I/carapace width 3.22- 3.35. Distribution.— Known only from the type locality on the eastern escarpment in central Madagascar (Fig. 98). Additional material examined. — MADAGAS- CAR: Toamasina Province, Parc National Perinet, near Andasibe, 18°56'S, 48°24'E, elev. 1000 m, 4- 5 November 1993 (C. Griswold) 2? (CAS). Alamnea new genus Type species.— merina new spe- cies. Etymology.— Combination of Malagasy Ala, and Latin Aranea, both meaning spiders, considered feminine. Diagnosis.— Anterior portion of abdomen of both sexes forming a sclerotized, annulate petiole produced dorsally into a short projec- tion or horn (Figs. 4, 66, 94); abdomen of males with a thin, shiny transparent dorsal scutum (Fig. 95); PP present (Figs. 53, 57, 73, 89). Description.— Total length 1.60-3.00. Car- apace of most species narrowly trapezoidal in dorsal view (Figs. 63, 66, 95, 96), oval in A. aiba new species (Fig. 69), length 1.39-1.67 times width, low, maxirnum height 0.38-0.52 width; texture finely rugose (Fig. 50), thoracic fovea a small, round pit, carapace posterior margin weakly concave medially, forming weakly upturned lip; ocular area with PFR width 1.83-2.56 times OAL, 2.14-2.69 times OQP, OQP 0.81-1.11 times OQA; diameter AM 1.00-1.60 times PM, distance PM-PL 0.80-1.50 times PM diameter; clypeal height 1.11-2.40 times AM diameter, cheliceral length 1.93-3.80 times clypeal height (Fig. 51); chelicerae unmodified. Sternum rugose (Fig. 49) to pustulate, length 0.88-1.15 times width, plural and sternal sclerotizations extend between and surround coxae (Figs. 2, 68, 94). Abdomen sclerotized from epigastric furrow to and surrounding pedicel (Figs. 63-69), sclerotization forming annulate petiole pro- duced dorsally into a short projection or horn (Figs. 4, 94), anterior sclerotization much broader in males, males with a thin, shiny transparent dorsal scutum (Fig. 95), abdomen otherwise unsclerotized, oval to triangular; ab- dominal setae short, slender, bases of anterior setae unmodified. Legs short, femur I length 1.63-2.11 times carapace width, unmodified (Figs. 68, 94). Male palpus (Figs. 57-62) with cymbial RMP pointing ventrad, smaller than PC; palpal bulb with dentate TL, apex a small, smooth to pustulate lobe; C median, longitu- dinal, simple or with accessory process, smooth or rarely dentate; E thick, making sim- 68 THE JOURNAL OF ARACHNOLOGY Figures 45-48. — Vazaha toamasina new species, holotype male, right palpus. 45. Retrolateral view; 46. Prolateral view; 47. Ventral view; 48. Cymbial base, retrolateral view. C = conductor; E = embolus; PC = paracymbium; RMP = retromedian cymbial process; TL = ventromedian tegular lobe. pie curve, origin apical between 10“ 11 o'clock, ridged; PP present, fleshy, pustulate, with or without teeth, thick or hooked apicaL ly; sperm duct with tight double twist (curli- cue) near embolic base. Epigynum (Figs. 55, 56, 74=77) with S and long MH with slender MS between CO, ML parallel-sided. Vulva (Figs. 90=93) with sclerotized, simple hemi- spherical lateral AD, in most specimens larger than HS, FD posterior. Composition.—Four species. Distribution. — Madagascar (Fig. 98). Alaranea betsileo new species Figs. 2=4, 6=14, 49=54, 56, 59=64, 74, 75, 90, 98 Types. — -Male holotype and female para- type from Madagascar, Fiaearantsoa Province, Parc National Ranomafana, Talatakely, mon- tane rain forest, 2ri5'S, 47°25^E, elev. 900 m, 5=7 November 1993 (C. Griswold) (CAS). Etymology. ^Named for the indigenous people of Fianarantsoa Province. Diagnosis. — -Conductor bipartite, with thin, broad proximal piece separate from C proper GRISWOLD— MADAGASCAR CYATHOLIPIDAE 69 Figures 49-52. — -Morphology of Alaranea betsiieo new species, from Talatakely, 49. Carapace, ventral view; 50. Carapace, dorsal view; 51. Carapace, anterior view; 52. Abdomen, ventral view; 49, 50. Male; 51, 52. Female. (Figs. 53, 61); in both sexes sternum dark red- brown to black, abdomen of most specimens with extensive dark markings, dorsum lacking sinuate longitudinal dark bands, with median black band surrounding 1-2 anterior white spots (Figs. 63, 64). Description.— Mafe (7 km W Ranoma- fana): Total length 2.24. Carapace dark red- brown, unmarked, ocular area with diffuse dark grey surrounding AM, black surrounding AL-PL; clypeus dusky grey in center, chelic- erae red-brown, with faint dark anterobasal streaks; palpal coxae red-brown, lighter at tips; labium and sternum dark brown to nearly black; coxae, legs, and palpi yellow-white, un- marked, cymbium dusky yellow-brown; ab- domen black dorsally beneath shiny transpar- ent scutum, dark transverse bands extending laterally from midpoint and posterior, those in middle nearly meeting ventrally, dark brown sclerotization extending from epigastric fur- row to and surrounding pedicel to form an- nulate petiole, sclerotization very broad anter- odorsally. Carapace 1.04 long, 0.64 wide, 0.26 high, trapezoidal in dorsal view; PER 038 wide, AER 0.39 wide, OAL 0.19; ratio AM: 70 THE JOURNAL OF ARACHNOLOGY Figures 53-58. — Genitalia of Alaranea spp. 53, 54, 57, 58. Left male palpus; 55, 56. Epigyeum; 53, 55-57. Ventral view; 54, 58. Retrolateral view; 53, 54, 56. Alaranea betsileo new species, from 7 km W Ranomafana; 55. A. alba new species, paratype; 57, 58. A, alba new species, holotype. AL:PM:PL, 1.33:1.0:1.0:1.0, PM diameter 0.06. Clypeus 0.14 high, chelicerae 0.33 long. Sternum 0.50 long, 0.50 wide; labium 0.10 long, 0.14 wide; palpal coxae 0.16 long, 0.12 wide. Leg measurements (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 2.60 + 0.42 + 2.08 + 2.20 + 1.36 = [8.66]; II: 2.32 + 0.40 + 2.32 + 1.96 + 1.28 = [8.28]; GRISWOLD— MADAGASCAR CYATHOLIPIDAE 71 Figures 59=62= — Alaranea betsileo new species, male from Talatakely, right palpus. 59. Retrolateral view; 60. Prolateral view; 61. Ventral view; 62. Parembolic process. A = apical lobe of tegulum; C = conductor; E = embolus; PP = parembolic process; ST = subtegulum; T = tegulum; TL = ventromedian tegular lobe. Ill: 1.48 + 0.36 + 1.48 + 1.04 -f 0.80 - [5.16]; IV: 2.20 + 0.40 + 1.72 + 1.56 + 0.88 = [6.76]; Palp: 0.28 + 0.10 + 0.09 + (absent) + 0.24 - [0.71]. Palp (Figs. 53, 54, 59-62) with cymbial RMP short, acutely pointed, with distal blunt projection, PC broad in lat- eral view; tegulum apex pustulate, TL large, convex, dentation extensive; C large, double, with flattened translucent lower article nearly as large as C proper; PP with apical recurved hook. Variation: {n — 3). Total length 2.12-2.61; ratios of carapace length/width 1.54-1.62, height/width 0.39-0.42; ratios of PER/OQP 2.37-2.53, PER/OAL 1.85-2.16, OQP/OQA 0.88-0.94 distance PM-PL/diameter PM 1.14-1.40, diameter AM/PM 1.00-1.60; ratios of clypeal height/AM diameter 1.62- 1.86, cheliceral length/clypeal height 2.36- 2.46; ratio of sternum length/width 0.88-1.00; ratio of length femur I/carapace width 1.87- 2.07. Markings of carapace dark brown to nearly black; legs pale yellow white to dusky gray; abdomen with dorsolateral transverse 72 THE JOURNAL OF ARACHNOLOGY marks entire to broken to rarely absent, dorsal black area ranges from narrow median band to totally covering dorsum. Female (7 km. W Ranomafana): Total length 2.24. Carapace red-brown, dusky on pars cephalica and around thoracic fovea, with median yellow-brown area between thoracic fovea and posterior margin of pars cephalica, ocular area dark grey between AM-PM and AL-PL, dusky marking extending below AM to lower margin of clypeus; chelicerae and palpal coxae yellow-brown, unmarked; labium and sternum dark brown to black; coxae, legs, and palpi yellow-white, unmarked; abdomen yellow-white, dorsum with broad longitudinal black mark, this mark forming lateral trans- verse bands near middle, small black spot at posterior apex, venter yellow-white, un- marked; dark brown sclerotization extending from epigastric furrow to and surrounding pedicel to form annulate petiole, sclerotization much less extensive anteriorly than in male. Carapace 0.94 long, 0.60 wide, 0.25 high; PER 0.37 wide, AER 0.36 wide, OAL 0.18; ratio AM:AL:PM:PL, 1.6: 1.2: 1.0: 1.2, PM di- ameter 0.06. Clypeus 0.11 high, chelicerae 0.28 long. Sternum 0.50 long, 0.47 wide; la- bium 0.10 long, 0.16 wide; palpal coxae 0.17 long, 0.12 wide. Leg measurements (femur + patella + tibia + metatarsus + tarsus = [To- tal]): I: 2.24 + 0.44 + 1.80 + 1.72 + 1.06 = [7.26]; II: 2.00 + 0.42 + 1.60 + 1.48 + 1.02 = [6.52]; III: 1.24 + 0.36 + 0.92 + 0.84 + 0.76 - [4.12]; IV: 1.84 + 0.40 + 1.36 + 1.20 + 0.78 = [5.58]; Palp: 0.21 + 0.09 + 0.11 + (absent) + 0.22 ~ [0.63]. Epigynum as in Figs. 56, 74, 75; vulva as in Fig. 90. Variation: {n — 3). Total length 2.18-2.89; ratios of carapace length/width 1.56-1.69, height/width 0.34-0.43; ratios of PER/OQP 1.91-2.30, PER/OAL 1.28-2.00, OQP/OQA 0.80-1.00, distance PM-PL/PM diameter 1.28-2.00, diameter AM/PM 1.14-1.67; ratios of clypeal height/AM diameter 1.40-2.00, cheliceral length/clypeal height 2.22-3.71; ra- tio of sternum length/width 1.03-1.10; ratio of length femur I/carapace width 1.92-2.17. Car- apace of most specimens dark brown to black, in rare specimens orange-brown, unmarked; dorsal abdominal dark markings range from narrow, broken laterally (Fig. 63) to broad, al- most obscuring dorsum (Fig. 64), anterome- dian white spot rarely obscure. Natural history. — Common inside forest hanging beneath sheet webs between 0.2-2 m above ground. Distribution. — Known only from montane forests near Ranomafana in Fianarantsoa Province (Fig. 98). Material examined. — MADAGASCAR; Fian- arantsoa Province: Parc National Ranomafana, Tal- atakely, montane rain forest, 21°15'S, 47°25'E, elev. 900 m, 34(3749 (including holotype and paratype), 5-7 November 1993 (N. Scharff, S. Larcher, C. Griswold, and R. Andriamasamanana) (one pair in MRAC, remainder divided among CAS, USNM, and ZMUC). Parc National Ranomafana, Vohipar- ara, 21°14'S, 47°24'E, elev. 900 m, 6369, 5-7 De- cember 1993 (N. Scharff, S. Larcher, C. Griswold, and R. Andriamasamanana) (CAS, ZMUC, USNM); Parc National Ranomafana, 200 m N re- search Cabin, trail G, beating, 132 9 (CAS) 232 9 (MCZ), 23 March 1992 (S. Kariko, V. Roth); Parc National Ranomafana, beating in forest, 1349 (CAS), 4312 9 (MCZ), 25 March 1992 (Emile); Parc National Ranomafana, 200 m N research Cab- in, trail G, beating, 2349, 25 March 1992 (B. Roth) (CAS); Parc National Ranomafana, 2ri2'S, 47°27'E, from foliage, elev. 1000 m, 49 (CAS) 3 9 (MCZ) April 1992 (V. & B. Roth); Parc National Ranomafana, 7 km W Ranomafana, elev. 900 m, 21°12'S, 47°27'E, 23, 20-24 March 1990, elev. 1100 m, 1319, 8-21 October 1988, 13, 21-30 Oc- tober 1988, 23, 1-7 November 1988 (W. Steiner) (USNM). Alaranea merina new species Figs. 5, 65, 66, 70-73, 76, 77, 91, 98 Types. — Male holotype and female para- type from Madagascar, Toamasina Province, Parc National Perinet, near Andasibe, 18°56'S, 48°24'E, elev. 1000 m, montane rain forest, 4-5 November 1993 (C.E. Griswold) (CAS). Etymology. — Named for the indigenous people of Antananarive Province. Diagnosis. — Conductor simple, proximal point narrower than cup (Figs. 70-72); dor- sum of abdomen with sinuate longitudinal dark bands diverging from apex to middle and converging posteriorly (Figs. 65, 66). There seem to be no consistent characters to separate females of merina new species from ardua new species, though in the former the cara- pace is pale yellow-brown with darker mark- ings along the borders of the pars cephalica (Fig. 66), whereas the carapace of ardua tends to be darker (Fig. 96). Description. — Male (7 km. W Ranoma- fana): Total length 2.32. Carapace yellow- GRISWOLD— MADAGASCAR CYATHOLIPIDAE 73 Figures 63-69.- — Morphology of Alaranea spp. 63-66, 69. Dorsal view, 64 and 65 abdomen only; 67. Ventral view; 68. Lateral view; 63, 64. Alaramea betsileo new species, females from Talatakely; 65, 66. Alaranea menna new species, females from Perinet; 67-69. Alaranea alba new species, hoiotype male. (Scale bar = 1 mm) brown, brown along margins of pars cephalica and on thoracic fovea; ocular area dark grey beginning just anterior of PER, black between AM and AL-PL; clypeus yellow-brown, dark grey in center from AM to clypeal margin; chelicerae and palpal coxae orange-brown; sternum and labium black; legs and palpi yek low-white, unmarked, cymbium yellow- brown, teguium orange-brown; abdomen white, with brown sclerotization extending from epigastric furrow to and surrounding pedicel to form annulate petiole, dorsum with faint longitudinal brown bands beneath trans- parent scutum, with dorsolateral elongate black spot and posterior lateral wavy line, posterior apex with black spot, venter dark gray between epigastric furrow and spiracle. Carapace LOO long, 0.64 wide, 0.28 high, trapezoidal in dorsal view; PER 0.39 wide, AER 0.38, OAL 0.19; ratio AM:AL:PM:PL 1.5:1.17:1.0:1.17, PM diameter 0.06. Clypeus 0.12 high, chelicerae 0.30 long. Sternum 0.50 74 THE JOURNAL OF ARACHNOLOGY Figures 70-73. — Alaranea merina new species, male from Perinet, right palpus. 70. Retrolateral view; 71. Prolateral view; 72. Ventral view; 73. Parembolic process. PC = paracymbium; RMP = retromedian cymbial process. long and wide; labium 0.12 long, 0.15 wide; palpal coxae 0.18 long, 0.14 wide. Leg mea- surements (femur + patella + tibia + meta- tarsus + tarsus = [Total]): I: 2.44 + 0.44 + 1.96 + 1.84 + 1.08 = [7.76]; II: 2.12 + 0.44 + 1.72 + 1.72 + 0.96 = [6.96]; III: 1.30 + 0.38 + LOO + 0.96 + 0.64 = [4.28]; IV: 2.00 + 0.40 + 1.48 + 1.40 + 0.72 = [6.00]; Palp: 0.32 + 0.14 + 0.10 + (absent) + 0.26 - [0.82]. Palp (Figs. 70-73) with bulb marked as in Alaranea betsileo new species (Figs. 53, 54), cymbial RMP simple, pointed, PC slender in lateral view; tegulum apex strongly pustu- late, TL large, projecting ventrally to form blunt point, denticulate over large area; C small, single; PP with apical recurved hook. Variation: {n = 3). Total length 2.29-2.71; ratios of carapace length/width 1.53-1.67, height/width 0.34-0.47; ratios of PER/OQP 2.18-2.44, PER/OAL 2.00-2.09, OQP/OQA 0.82-1.00, distance PM-PL/diameter PM 1.00-1.67, diameter AM/PM 1.25-1.67; ratios of clypeal height/AM diameter 1.36-1.60, cheliceral length/clypeal height 2.55-3.28; ra^ tio of sternum length/width 1.03-1.1 1; ratio of length femur I/carapace width 1.89-2.12. Car- GRISWOLD— MADAGASCAR CYATHOLIPIDAE 75 Figures 74-77. — Epigyna of Alaranea spp. 74, 76. Ventral view; 75, 77. Lateral view; 74, 75. Alaranea betsileo new species, Talatakely; 16-11 . Alaranea merina new species, Perinet. apace with or without faint dusky radii ex- tending from thoracic fovea; abdominal dor- sum (Figs. 65, 66) clear with dorsolateral markings visible to obscured to varying de- grees by black, these markings range from median transverse band or U to large dark area, lateral black marks present or absent. Female (7 km. W Ranomafana): Total length 2.28. Markings as in male except ab- domen having dorsomedian brown bands fainter, lateral black spots larger, and posterior spot smaller. Carapace 0.92 long, 0.58 wide, 0.24 high; PER 0.33 wide, AER 0.36 wide, OAL 0.18; ratio of eyes AM:AL:PM:PL: 1.6: 1.2: 1.0: 1.2, PM diameter 0.06. Clypeus 0.11 high, chelicerae 0.28 long. Sternum 0.46 long, 0.48 wide; labium 0.1 1 long, 0.14 wide; palpal coxae 0.17 long, 0.12 wide. Leg measure- ments (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 2.24 + 0.44 + 1.80 T 1.72 + 1.06 = [7.26]; II: 2.00 + 0.42 + 1.60 + 1.48 + 1.02 = [6.52]; III: 1.24 + 0.36 + 0.92 + 0.84 + 0.76 = [4.12]; IV: 1.84 + 0.40 + 1.36 + 1.20 + 0.78 = [5.58]; Palp: 0.21 + 0.09 + 0.11 + (absent) + 0.22 - [0.63]. Epi- gynum and vulva as in Alaranea betsileo new species, epigynum as in Figs. 76, 77; vulva as in Fig. 91. Variation: (n = 4). Total length 2.00-2.82; ratios of carapace length/width 1.55-1.60, height/width 0.31-0.39; ratios of PER/OQP 2.09-2.55, PER/OAL 1.83-2.44, OQP/OQA 0.86-1.00, distance PM-PL/diameter PM 0.88-1.50, diameter AM/PM 1.25-1.67; ratios of clypeal height/AM diameter 1.10-1.50, cheliceral length/clypeal height 2.83-3.67; ra- tio of sternum length/width 1.03-1.15; ratio of length femur I/carapace width 1.87-2.17. Car- 76 THE JOURNAL OF ARACHNOLOGY apace yellow-brown to orange-brown, may be darker along margins of pars cephalica; ab- dominal dorsum with faint longitudinal brown bands exposed (Fig. 65) or obscured by small to large dorsolateral black spot (Fig. 66), may have posterior lateral dark spot or wavy line. Natural history. — Common inside forest hanging beneath sheet webs between 0.2-2 m above ground. Distribution. — Widespread in mid-eleva- tion forests along the eastern side of the es- carpment (Fig. 98). Material examined. — MADAGASCAR; Fian- arantsoa Province: 43 km. S Ambalavao, Reserve Andringitra, 22°14'S, 47°00'E), elev. 825 m, sifted litter, rainforest, M, 5 October 1993 (B. L. Fisher) (CAS); Massif Andringitra, Mahasoa, elev. 2100 m, 19, October 1971 (B. Ranson) (MRAC); Parc Na- tional de Ranomafana: around research cabin, 2c33 9 , 26 March 1992 (V. & B. Roth, S. Kariko) (MCZ). Parc National de Ranomafana, from foliage, ca. 2ri2'S, 47°27'E, elev. ca. 1000 m, 1(33 9, April 1992 (V. & B. Roth, S. Kariko) (CAS); 7 km. W Ranomafana, elev. 1100 m, 19, 22-31 October 1988, 2(33 9, 1-7 November 1988 (W.E. Steiner) (USNM); Elev. 1200 m, 19, 22 October 1988 (W. Steiner, C. Kremen, R. Van Epps) (USNM); Parc Na- tional de Ranomafana, Vohiparara, ca. 21°14'S, 47°24'E, elev. 1100 m, 49, 5-7 November 1993 (N. Scharff, S. Larcher, C. Griswold, R. Andriamasa- manana) (CAS, USNM, ZMUC). Parc National de Ranomafana, Talatakeley, 21°15'S, 47°25'E, elev. 900 m, 7(3219, 5-7 December 1993 (C. Griswold, N. Scharff, S. Larcher, and R. Andriamasamanana) (CAS, USNM, ZMUC). Toamasina Province: Parc National Perinet, near Andasibe, 18°56'S, 48°24'E, elev. 1000 m, montane rain forest, 40(3 30 9 , 4-5 No- vember 1993 (J. Coddington, S. Larcher, C. Gris- wold, R. Andriamasamanana, & N. Scharff)(CAS, USNM, ZMUC); Perinet, 18°55'S, 48°25'E, 19,1- 3 August 1992 (V. & B. Roth) (CAS); Foret de Didy, arbustes, 1(3, March 1947 (MNHN); Mandraka, bat- tage, 3(37 9, December 1946 (J. Millot) (MNHN); Beanana, 15°44'S, 49°28'E, 1(3, February 1970 (A. Lambillon)(MRAC). Alaranea alba new species Figs. 55, 57, 58, 67-69, 78-83, 92, 98 Types. — Male holotype and 1(339 para- types from Beria, Madagascar, June 1969 (A. Lambillon) (MRAC 142.978), MRAC except 1(319 (CAS). Etymology. — The species name refers to the largely white coloration. Diagnosis. — Conductor undivided, proxi- mal point elongate attenuate (Figs. 58, 79, 80); sternum pale yellow-brown, abdomen white marked only with lateral, ventral, and posterior black spots (Figs. 67-69). Description. — Male (holotype): As in Figs. 67-69. Total length 1.80. Carapace pale yel- low-brown, unmarked, thoracic fovea brown, ocular area black on ocular quadrangle and between lateral eyes; clypeus, chelicerae, ster- num, labium, and palpal coxae yellow-brown, unmarked; legs and palpi yellow-white, un- marked; cymbium and tegulum yellow-brown; abdomen white, with brown sclerotization extending from epigastric furrow to and sur- rounding pedicel to form annulate petiole, dorsum with faint dorsolateral dusky markings beneath shiny transparent scutum, with black oval spots laterally, ventrally, and at posterior apex. Carapace 0.86 long, 0.58 wide, 0.30 high, oval in dorsal view; PER 0.35 wide, AER 0.34, OAL 0.16; ratio AM:AL:PM:PL, 1.2: 1.0: 1.0: 1.5, PM diameter 0.05. Clypeus 0.10 high, chelicerae 0.32 long. Sternum 0.44 long, 0.42 wide; labium 0.08 long, 0.12 wide; palpal coxae 0.16 long, 0.10 wide. Leg mea- surements (femur + patella + tibia -f meta- tarsus + tarsus = [Total]): I: 1.92 + 0.36 + 1.48 + 1.44 + 0.96 = [6.16]; II: 1.72 + 0.36 + 1.44 + 1.36 -f 0.88 = [5.76]; III: 1.12 + 0.28 + 0.72 + 0.88 + 0.56 = [3.56]; IV: 1.60 + 0.30 + 1.20 + 1.04 + 0.64 = [4.78]; Palp: 0.28 + 0.10 + 0.08 + (absent) T 0.28 = [0.74]. Palp (Figs. 57, 58, 78-81) with cym- bial RMP bifid, with outer ventrad- and inner distad-directed processes, PC slender, pointed in lateral view; tegulum apex weakly pustu- late, TL pointed ventrally, denticulate area small; C large, single, complex, with prolater- al smooth concavity and retrolateral slender basad-directed process; PP large, swollen, with recurved apical process. Variation: (n = 2). Total length 1.80-1.84; ratio of carapace height/width 0.40-0.52; ra- tios of PER/OQP 2.54-2.69, PER/OAL 2.19- 2.36, OQP/OQA 0.93-1.00, distance PM-PL/ diameter PM 1.40-1.50, diameter AM/PM 1.20-1.50; ratios of clypeal height/AM di- ameter 1.67-2.33, cheliceral length/clypeal height 1.93-3.20; ratio of sternum length/ width 0.91-1.04; ratio of length femur I/car- apace width 1.63-1.65. Female (paratype): Total length 1.72. Markings and structure as in male, except sclerotization of abdominal petiole weaker, yellow- white. Carapace 0.82 long, 0.54 wide. GRISWOLD— MADAGASCAR CYATHOLIPIDAE 77 Figures 78-81. — Alaranea alba new species, holotype male, right palpus. 78. Retroventral view; 79. Proventral view; 80. Ventral view; 81. Cymbial base, retroapical view. PC = paracymbium; RMP retromedian cymbial process. 0.26 high; PER 0.30 wide, AER 0.29 wide, OAL 0.12; ratio AM:AL:PM:PL, 1.5: 1.0: 1.0: 1.0, PM diameter 0.05. Clypeus 0.10 high, chelicerae 0.24 long. Sternum 0.40 long and wide; labium 0.08 long, 0.12 wide; palpal coxae 0.12 long, 0.10 wide. Leg measure- ments (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 2.08 + 0.36 + 1.76 + 1.68 + 1.12 - [7.00]; II: 2.00 + 0.36 + 1.64 + 1.52 + 0.96 = [6.48]; III: 1.16 + 0.28 + 0.72 + 0.84 + 0.60 - [3.60]; IV: 1.80 + 0.32 + 1.40 + 1.20 + 0.80 = [5.52]; Palp: 0.20 + 0.08 + 0.12 + (absent) + 0.20 = [0.60]. Epi- gynum as in Figs. 55, 82, 83, distance from tip of S to posterior margin greater than in Alaranea betsileo new species; vulva as in Fig. 92, hemispherical AD relatively larger in relation to HS than in other Alaranea. Variation: (n = 3). Total length 1.64-1.84; ratios of carapace/width length 1.39-1.52, height/width 0.38-0.48; ratios of PER/OQP 2,14-2.50, PER/OAL 2.50-2.58, OQP/OQA 0.86-1.00, diameter AM/PM 1.00-1.20; ratios of clypeal height/ AM diameter 1.67-2.40, cheliceral length/clypeal height 2.50-2.60; ra- tio of sternum length/width 0.95-1.05; ratio of length femur Ecarapace width 1.78-1.92. Distribution.— Known only from the type locality near Beria at 19°40'S, 45°23'E, in To- liara Province, Madagascar (Fig. 98). 78 THE JOURNAL OF ARACHNOLOGY Figures 82-85. — Epigyna of Alaranea spp. 82, 84. Ventral view; 83, 85. Lateral view; 82, 83. Alaranea alba new species, paratype; 84, 85. Alaranea ardua new species, Marojejy. Material examined. — Only the type series. Alaranea ardua new species Figs. 84-89, 93-98 Types. — Male holotype and female para- type from Madagascar, Antsiranana Province, Marojejy Reserve, 8.4 km NNW Mananteni- na, montane rain forest, 14°26'S, 49°45'E, elev. 700 m, 10-16 November 1993, C. Gris- wold (CAS). Etymology. — The species name is from the Latin for difficult, hard-won. Diagnosis.— Conductor simple, proximal point thick, bifid, equal in width to cup (Figs. 86-88); dorsum of abdomen with sinuate lon- gitudinal dark bands diverging from apex to middle and converging posteriorly (Figs. 95, 96). There seem to be no consistent characters to separate females of ardua from merina. though the carapace of ardua (Fig. 96) tends to be darker than that of merina (Fig. 66). Description. — Male (holotype): Total length 2.79. Carapace (Fig. 95) dusky orange- brown, faintly mottled with grey, especially along lateral margin, small dark longitudinal band anteriad of thoracic fovea; ocular area black surrounding AM and lateral eyes, ocular quadrangle dark grey; clypeus yellow-brown, dark grey mark beneath AM narrowing to clypeal margin; chelicerae orange-brown, with faint dark basal streaks; sternum, labium, and palpal coxae red-brown with dark mottling, sternum black along ridges of rugosity, ap- pearing nearly black; coxae, legs and palpi white, unmarked, palpal tibia yellow-brown, cymbium dark red-brown (Fig. 94); abdomen white, dorsum (Fig. 95) with paired longitu- dinal dark grey bands beneath transparent GRISWOLD—MADAGASCAR CYATHOLIPIDAE 79 Figures 86-89.— Ahmnea ardua new species, male from Marojejy, right palpus. 86. Retrolateral view; 87. Prolateral view; 88. Ventral view; 89. Parembolic process. shiny scutum, area between these bands dusky, sides and posterior apex with black spots, venter grey between epigastric furrow and spiracle, dark brown sclerotizatioe ex- tending from epigastric furrow to and sur- rounding pedicel to form annulate petiole. Carapace 1.24 long, 0.76 wide, 0.35 high, trapezoidal in dorsal view; PER 0.47 wide, AER 0.44 wide, OAL 0.21; ratio AM:AL:PM: PL, 1.28:1.14:1.0:1.14, PM diameter 0.07. Clypeus 0.14 high, chelicerae 0.34 long. Ster- num 0.59 long, 0.58 wide; labium 0.13 long, 0.18 wide; palpal coxae 0.21 long, 0.16 wide. Leg measurements (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 1.53 + 0.28 + 1.28 + 1.13 + 0.70 = [4.92]; II: 1.51 + 0.25 + 1.17 + 1.04 + 0.64 - [4.61]; III: 0.98 + 0.23 + 0.64 + 0.59 + 0.38 = [2.82]; IV: 1.28 + 0.21 + 0.96 + 0.85 + 0.45 = [3.75]; Palp: 0.18 + 0.07 + 0.05 + (absent) + 0.18 = [0.48]. Palp (Figs. 86-89) with bulb marked as in Aiamnea betsileo new species, cymbial RMP very short, acute, PC slender in lateral view; tegulum apex pustulate, TL large, con- vex, denticuiation extensive; C large, retrola- terally dentate, with projecting basal article; PP with apical recurved hook. Variation: (n = 3). Total length 2.57-3.00; 80 THE JOURNAL OF ARACHNOLOGY Figures 90-93. — Vulvae of Alaranea spp., cleared, dorsal view. 90. Alaranea betsileo new species, from 7 km W Ranomafana; 91. Alaranea merina new species, Mandraka; 92. Alaranea alba new species, paratype; 93. Alaranea ardua new species, Marojejy. AD = afferent duct; FD = fertilization duct; HS = spermathecal head. ratios of carapace length/width 1.60-1.64, height/width 0.45-0.46; ratios of PER/OQP 2.25-2.55, PER/OAL 2.00-2.25, OQP/OQA 0.81-1.11, distance PM-PL/diameter PM 1.00-1.43, diameter AM/PM 1.14-1.57; ratios of clypeal height/AM diameter 1.44- 1.62, cheliceral length/clypeal height 2.19- 3.30; ratio of sternum length/width 1.02-1.07; ratio of length femur I/carapace width 1.90- 2.15. Markings of carapace range from dusky orange-brown to dark brown, dorsum of ab- domen with longitudinal dark markings nar- row and separate (Fig. 95) to completely black beneath scutum, lateral transverse marks forming spot or band connected to dorsum. Female (paratype): Total length 2.74. Markings (Figs. 96, 97) as in male except che- licerae, abdomen, and palpal coxae dark red- brown, sternum and petiole black, dorsum of abdomen with broad median black mark, this extending anteriad to sclerotized petiole in two bands surrounding white mark, and ex- tending laterally to form median transverse band, posterior tip black, venter pale. Cara- pace 1.13 long, 0.71 wide, 0.31 high; PER 0.45 wide, AER 0.43 wide, OAL 0.20; ratio AM:AL:PM:PL, 1.5:1.33:1.0:1.17, PM diam- eter 0.06. Clypeus 0.11 high, chelicerae 0.35 long. Sternum 0.56 long, 0.48 wide; labium 0.11 long, 0.19 wide; palpal coxae 0.21 long, 0.13 wide. Leg measurements (femur + pa- tella + tibia + metatarsus + tarsus = [Total]): I: 1.34 + 0.28 + 1.15 T 1.23 + 0.64 - [4.64]; II: 1.28 + 0.25 + 1.21 + 1.15 + 0.64 = [4.53]; III: 0.76 + 0.21 + 0.55 + 0.51 + 0.40 = [2.43]; IV: 1.17 T 0.21 + 0.87 + 0.74 + 0.47 = [3.46]; Palp: 0.26 + 0.08 T 0.14 T (absent) + 0.26 = [0.74]. Epigynum and vul- va as in Alaranea betsileo new species, epi- gynum as in Figs. 84, 85; vulva as in Fig. 93. Variation: (n = 3). Total length 2.32-3.46; ratio of carapace height/width 0.45-0.52; ratios of PER/OQP 2.16-2.39, PER/OAL 2.16-2.26, OQP/OQA 0.90-1.00, distance PM-PL/diam- eter PM LOO-1.43, diameter AM/PM 1.28- 1.50; ratios of clypeal height/AM diameter 1.11-1.40, cheliceral length/clypeal height 3.00-3.80; ratio of sternum length/width 1.02- 1.15; ratio of length femur I/carapace width 1.98-2.20. Markings of carapace range from orange except black ocular area to all dark brown; dorsal abdominal markings range from faint to bold, dorsolateral bands may be narrow and broken, solid and separate (Fig. 96) or meeting medially, or entirely black. Natural History. — Common inside forest Figures 94-97. — Morphology of Alaranea ardua new species, from Marojejy. 94, 95. Male; 96, 97. Female; 94. Lateral view; 95, 96. Dorsal view; 97. Ventral view. (Scale bar = 1 mm) hanging beneath sheet webs between 0.2-2 m above ground. Distribution. — Known only from the type locality (Fig. 98). Material examined. — MADAGASCAR: Antsi- ranana Province, Marojejy Reserve, 8.4 km NNW Manantenina, montane rain forest, 14°26'S, 49°45'E, elev. 700 m, 10-16 November 1993 (J. Coddington, N. Scharff, S. Larcher, C. Griswold, and R. Andriamasamanana) 13(3119 (CAS, ZMUC, USNM). DISCUSSION So far as is known, Malagasy cyatholipids occur in moist forest, the majority being re- corded from above 600 m elevation along the eastern slopes of the central mountain chain (Fig. 98). At least Alaranea merina new spe- cies occurs at over 2000 m. Ulwembua antsi- ranana new species, which occurs in an area of local orographic rainfall, is disjunct from the main distribution of Cyatholipidae. Re- striction to moist forest appears likely. Col- lecting by the author and colleagues in drier habitats never revealed Cyatholipidae. At least two Malagasy genera, Ulwembua and Alaranea, show affinities to taxa occur- ring in tropical or subtropical montane forests of eastern Africa. Ulwembua was previously known from three species from South Africa 82 THE JOURNAL OF ARACHNOLOGY Figure 98. — Map showing distributions of Cy- atholipidae in Madagascar. (Griswold 1987): U. outeniqua Griswold from the coastal forests of central Cape Province, and U. pulchra Griswold and U. denticulata Griswold from Zululand. Undescribed species of Ulwembua also occur in the mountains of Tanzania. Alaranea new genus shows affini- ties to an undescribed genus occurring in montane forests from Malawi to Kenya. The distribution of the sister groups of these Mal- agasy cyatholipids is consistent with the Af- romontane biogeographic pattern detailed for spiders by Griswold (1991) in which the sister area of Madagascar comprises the tropical montane forests of the eastern part of Africa. Several groups of spiders, including Phyxelida and the Lamaika group of the Amaurobiidae Phyxelidinae (Griswold 1990) and Ulwembua and Alaranea of the Cyatholipidae, show this intercontinental disjunction, suggesting that their distribution is not the result of accidental dispersal. Their distribution may date from times of former connection or at least greater proximity between Madagascar and eastern Africa, perhaps in the Mesozoic (Rabinowitz et al. 1983). Phylogenetic and biogeographic evidence continues to support the suggestion that the Afromontane biota, at least the arach- nid component, is ancient. ACKNOWLEDGMENTS Principal support for this project was pro- vided by National Science Foundation grant DEB-9296271, with additional support from the Exline-Frizzell Fund, California Academy of Sciences, Scholarly Studies program of the Smithsonian, postdoctoral fellowships from the Smithsonian Institution and Kalbfleisch Fellowships from the American Museum of Natural History. Collection and export of specimens from Madagascar was made pos- sible through permits from the Association National Pour La Gestion des Aires Protegees (ANGAP) and Ministere des Eaux et Forets, expedited by the Xerces Society under their Accord de Collaboration with the Ministere des Universites, Republique de Madagascar. Dr. Claire Kremen and Mr. Cesaire Ramilison provided invaluable assistance. Rija Andri- masamanana, Roland Christophe, Jonathan Coddington, Scott Larcher, and Nikolaj Scharff collected cyatholipids and helped in the field. I particularly wish to thank Dr. Coddington of the Smithsonian Institution for hospitality provided while I worked in his lab. All habitus illustrations are by Jenny Speckels. Assistance with manuscript preparation was provided by Ms. Johanna Brandriff and Mr. Darrell Ubick; assistance with scanning electron microscopy was provided by Mrs. Susan Breydon and D. Ubick. A draft of the manuscript was read and criticized by Rudy Jocque, Wojciech Pulaws- ki, and Nikolaj Scharff. LITERATURE CITED Alderweireldt, M. & R. Jocque. 1994. Biodiversity in Africa and Europe: The Case of Spiders (Ara- neae). Biol. Jb. Dodonaea, 61:57-67. Brignoli, P.M. 1983. A catalogue of the Araneae described between 1940-1981. Manchester: Manchester Univ. Press, 755 pp. Coddington, J.A. 1983. A temporary slide mount allowing precise manupulation of small struc- tures. Verb. Naturwiss. Ver. Hamburg (NF), 26: 291-292. Davies, V.T 1978. A new family of spiders (Ara- neae: Teemenaaridae). Symp. Zool. Soc. London, 42:293-302. Forster, R.R. 1988. Cyatholipidae. Pp. 7-34, In Spiders of New Zealand, vol. 6. Otago Mus. Bull. Griswold, C.E. 1987. A review of the southern Af- GRISWOLD— MADAGASCAR CYATHOLIPIDAE 83 rican spiders of the family Cyatholipidae Simon, 1894 (Araneae: Araneomorphae). Ann. Natal Mus., 28:499-542. Griswold, C.E. 1990. A revision and phylogenetic analysis of the spider subfamily Phyxelidinae (Araneae, Amaurobiidae). Bull. American Mus, Nat. Hist., 196:1-206. Griswold, C.E. 1991. Cladistic biogeography of af- romontane spiders. Australian Syst. Bot., 4:73- 89. Jocque, R. 1994. Halidae, a new spider family from Madagascar (Araneae). Bull. British Arach- nol. Soc., 9:281-289. Merrett, R, G.H. Locket, & A.E Millidge. 1985. A check list of British spiders. Bull. British Arach- nol. Soc., 6:381-403. Merrett, R, & A.E Millidge. 1992. Amendments to the check list of British spiders. Bull. British Ar~ achnol. Soc., 9:4-9. Myers, N. 1988. Threatened biotas: “hot spots” in tropical forests. The Environmentalist, 8:187- 208. National Research Council. 1980. Research prior- ities in tropical biology. Natl, Acad. Sci., Wash- ington D.C., 116 pp. Rlatnick, N.I. 1979. [Review of] Arachnology. ed- ited by R Merrett. New York: Academic Rress, 1978, Symp. Zool. Soc. London, 42. Syst. ZooL, 28:115-117. Platnick, N.I, 1989. Advances in spider taxonomy: a supplement to Brignoli’s “A Catalogue of the Araneae described between 1940 and 1981”. Manchester Univ. Press, 673 pp. Platnick, N.I. 1993. Advances in spider taxonomy, 1988-1991: with synonymies and transfers 1940-1980. New York Entomol. Soc., 846 pp. Rabinowitz, RD., M.F. Coffin & D. Falvey. 1983. The separation of Madagascar and Africa. Sci- ence, 220:67-69. Rasoanaivo, R 1990. Rain forests of Madagascar: sources of industrial and medicinal plants. Am- bio, 19:421-424. Roewer, C.E 1942. Katalog der Araneae von 1758 bis 1940. Bremen: Natura, 1:1-1040. Simon, E. 1894. Histoire Naturelle des Araignees. 2nd ed. Raris: Roret, 1, Rp. 489-760. Wunderlich, J. 1978. Zur Kenntnis der Cyatholi- pinae Simon 1894 (Arachnida: Araneida: ?Te- tragnathidae). Zool. Beitr., 24:33-41. Wunderlich, J. 1993. Die ersten fossilen Becher- spinnen (Fam. Cyatholipidae) in Baltischem und Bitterfelder Bernstein (Arachnida: Araneae). Mitt. Geol.-Ralaont. Inst. Univ, Hamburg, 75: 231-241. Manuscript received I February 1996, revised 14 June 1996. 1997. The Journal of Arachnology 25:84-92 A NEW SPECIES OF SCHIZOCOSA FROM THE SOUTHEASTERN USA (ARANEAE, LYCOSIDAE) Gail E. Stratton: Department of Biology, Rhodes College, 2000 N. Parkway; Memphis, Tennessee 38112 USA. ABSTRACT. A new species of Schizocosa (Araneae, Lycosidae) is described and illustrated. Schizocosa uetzi new species is locally abundant in the southeastern USA and is mature during June and July. Morphological characters, coloration and courtship behavior separating this new species from its closely related congeners are noted. Members of the wolf spider genus Schizo- cosa Chamberlin 1904 from the Nearctic Re- gion were last revised by Dondale & Redner (1978). Since that time, two additional species have been described from the region by Uetz & Dondale (1979) and Stratton (1991). Within the genus there are at least two species groups that appear to be diverging by secondary sex- ual characteristics and by courtship behavior. The group best studied so far is the Schizo- cosa ocreata species group, defined by the presence of a finger-like paleal process on the male’s palp and by paired excavations on the transverse piece of the female’s epigynum. The group includes S. ocreata (Hentz 1844), S. crassipes (Walckenaer 1837), S. floridana Bryant 1934, S. rovneri Uetz & Dondale 1979, S. stridulans Stratton 1991 and S. uetzi new species. Uetz & Dondale (1979) de- scribed S. rovneri as the sister group to S. ocreata, the two differing only in the presence (in ocreata) or absence of the distinctive tibial bristles found on the first pair of legs; they are otherwise identical. Uetz & Denterlein (1979) described the courtship behavior of S, rovneri and demonstrated that courtship behavior serves as an isolating mechanism between S. ocreata and S. rovneri. This species and S. stridulans were first recognized as new spe- cies by differences in male secondary sexual characteristics (lack of pigment on tibia of males of S. rovneri, pigment but not bristles found on the tibia and distal portion of the femora in males of S. stridulans); and later, the courtship behavior in each was found to be distinct and to function as an isolating mechanism (Uetz & Denterlein 1979; Stratton 1991; in press). Schizocosa uetzi new species was first noted from collections from the southeastern USA made by W.R Maddison in 1984 and by the author and L. Williams in 1985. In each case, specimens were collected that keyed to S. ocreata, but secondary sexual characteristics did not match any of the known species. Subsequent work demonstrated that this species has a consistent pattern of sec- ondary sexual characteristics and distinctive courtship behaviors. METHODS Wolf spiders were collected throughout the southeastern USA during the springs and sum- mers of 1991-1995. Immature and mature in- dividuals were returned to the laboratory at the University of Mississippi where they were individually maintained in vials (8.5 cm X 5 cm) with wicks that extended into a water tray providing a constant source of moisture. Im- mature spiders were held until they matured, and mature spiders were held for behavioral studies. Appropriately- sized crickets were of- fered twice weekly as food for the spiders. Temperature in the laboratory ranged from 22-25 °C. Temperature during courtship and copulatory studies was 22-25 °C. Spiders were exposed to an L:D schedule of 14:10 h. Animals for behavioral studies were observed from within a few days to a few weeks of collection. Courtship and copulation were observed by setting the female in a culture dish with a piece of filter paper 6-12 h before observa- tions. Males and females were then placed in an observation chamber with the filter paper 84 STRATTON— A NEW SPECIES OF SCHIZOCOSA 85 where their interactions were videotaped using a Panasonic HD-5000 videocamera with a 105 mm macrolens. Sounds were recorded from the substrate by a stereo-needle transducer at- tached to an EG&G PARC, Model 113, preamp (Gain set at 5K, low roll off set at 0.3 Hz, high roll off at 10 kHz) and were over- layed onto videotape. Both courtship behav- iors and copulatory behaviors were videore- corded. Measurements were made of mature speci- mens (males, n = 22; females, n ^ 8) with an ocular micrometer. Terminology is as used in Dondale & Redner (1978) and Stratton (1991). The following abbreviations are used for collectors and for museum depositions: GES = Gail E. Stratton, PRM = Patricia R. Miller, GEM - Gary L. Miller, WRM - Wil- liam Miller, GTB = Gerry Baker, EAH = Ei- leen Hebets, WG ^ Wendy Garrison, KW = ICimball White, LEW - Eisa Wiliams, WPM = Wayne Maddison, TS = Terry Schiefer, ME = Micky Eubanks, KB == Kari Benson. Mu- seums: National Museum of Natural History Smithsonian Institution (USNM); American Museum of Natural History (AMNH); Cali- fornia Academy of Science (CAS); Mississip- pi Entomological Museum (MEM); Museum of Comparative Zoology (MCZ); Florida State Collection of Arthropods (ESC A); Biosyste- matics Research Institute, Canada (BRI); Field Museum of Natural History (FMNH). Speci- mens in the collection of the author are held at the University of Mississippi (UM). Schizocosa uetzi new species Figs. 1-6 Holotype.— -Male from USA, Mississippi, Eafayette County, 8 mi SE Oxford, TIOS R3W Sec. 35; 34°36'N, 89°29'W; 10 June 1991 (G. Stratton, PR. Miller, G.E. Miller, W.R. Miller, M. Eubanks), Illustrated speci- men. Collection #91-14; night, mixed pine, hardwood. Deposited in USNM. Paratypes.— MISSISSIPPI: Lafayette County, 8 mi SE Oxford, “Eonesome 80”, TIOS R3W Sec. 35; 34°36'N, 89°29'W; mixed pine hardwood, 1$, 10 June 1991 (G. Strat- ton, PR. Miller, G.E. Miller, WR. Miller, M. Eubanks), night, mixed pine hardwood (USNM); 2(3 (MEM); 26 (AMNH); 2(3 (FMNH); 2(3 (MCZ); 26 (CAS); 26 (BRI); 2(3 (FSCA). 28 June-5 July 1993, 2d (USNM). LOUISIANA: Natchitoches Parish, Kisatchie National Forest, “Red Dirt Area,” 10(3, coll. 23 May 1993, matured 7-25 June (GES, PRM) (UM), 89, matured 6 June-27 June (UM). Etymology. — The specific epithet is to honor Dr. George W. Uetz, spider ecologist, educator, mentor and friend. Diagnosis. — Males of Schizocosa uetzi new species possess a finger-like process on the palea of the palp as all other members of the ocreata group (Fig. 1; see also Dondale & Redner 1978, figs. 1-3; Stratton 1991, figs. 1- 4). S, uetzi can be distinguished from ocreata and crassipes by absence of a dense tibial brush, from rovneri by having patella and tib- ia I darker than the femur, and from stridulans in lacking a dark line on femur I and pigment on distal end of femur. Schizocosa uetzi is slightly, but significantly, larger than S. stri- dulans in total length, carapace length and carapace width {P < 0.5; r-test) (Table 1) and although it is not significantly different in size from S. rovneri (Table 1), S. uetzi has longer legs relative to its body size than S. rovneri {P < 0.5; Ntest) (Table 2). Table 3 summarizes the differences in genitalic dimensions be- tween these species. Schizocosa uetzi would probably not be confused with S. floridana, another member of this species group, as S. floridana is limited to Florida and has an un- dulating median band on the carapace that is different from all other members of this spe- cies group (Table 3). Females of S. uetzi new species can be confidently placed in the ocrea- ta species group by the presence of paired ex- cavations on the transverse piece of the epi- gynum. However, the specific determination can only be made when either S. uetzi new species is the only species in the ocreata group present in a habitat, or when a female is collected in copula with a male of S. uetzi. A key to adult males in the S. ocreata species group is provided. Description. — Males: (Figs. 1-3) {n = 22). For measurements see Table 1 (body length) and Table 2 (leg length). Cephalothorax brown; submarginal band narrow but distinct and wavy, sometimes with three spots on lat- eral sides; pale median band as wide as pos- terior lateral eyes with slight indentation at posterior one-third of the band. Sternum yel- low-brown to light brown with no spots, al- ways darker than coxae. Chelicerae brown, se- taceous, with two dark stripes down anterior 86 THE JOURNAL OF ARACHNOLOGY Figures 1-5. — Leg I of male and genitalia of males and females of Schizocosa uetzi new species (male holotype; female paratype). 1. Leg I of mature male; 2. Ventral aspect of left palp; 3. Enlargement of palp; 4. External aspect of epigynum of female; 5. Internal aspect of epigynum of female. IPE = intro- mittent portion of embolus; MA = median apophysis; TA = terminal apophysis, PPR = paleal process; MS = median septum; TP = transverse piece; EX = excavation of transverse piece; SP = spermatheca. side. Promargin of fang furrow with three un- evenly-sized teeth; retromargin of fang furrow with three evenly-sized teeth. Femora I-IV yellow to light brown with 3-4 dark annula- tions. Patellae I-IV brown. Tibia I brown to dark brown with black hairs, always slightly but distinctly darker than tibiae II-IV, always darker than femora I (Fig. 1), sometimes with faint annulations. Tibiae ILIV yellow to light brown. Tibial length to width ratio larger for males than females (Table 2). Dorsum of ab- domen in most specimens (16 of 21 speci- mens) with faint heart mark and with spots. Venter of abdomen yellow with black spots. Population from Louisiana, Natchitoches Par- ish, with dark square of pigment near genital pore (6 of 6 specimens). Palpal cymbium with 7-13 terminal macrosetae. Palpal palea (PPR) with long distal process, sometimes slightly curved into an .y-shape (Fig. 2). Median apophysis (MA) with distal margin convex. Intromittent part of embolus (IPE) slender, pointed with slight curve (Fig. 3). Terminal apophysis (TA) with thickened margin ex- tending to base of IPE. Length of palp, cym- bium and paleal process given in Table 3. File STRATTON— A NEW SPECIES OF SCHIZOCOSA 87 Table 1. — Comparison of total length, cephalothorax length and cephalothorax width of Schizocosa uetzi new species, S. rovneri (measurements of males from collections of GES; females from Uetz & Dondale 1979) and S. stridulans (data from Stratton 1991). Measurements are in mm, means are ± SD. Males are from type locality (n = 16) and from LA, Natchitoches Parish {n = 6); females from Natchi- toches Parish (n — 8). For any one characteristic measured, significant differences are indicated by different letters. Measurements that are followed by the same letter are not significantly different from each other (Students’ t-test, two tailed, P < 0.05). Schizocosa Schizocosa Schizocosa uetzi stridulans rovneri Males Total length (mean) 7.16 ±0.58 A 6.40 ± 0.43 B 6.78 ± 0.65 A (range) 5.9-8.0 5.04-6.80 6.0-7.8 Cephalothorax length (mean) 3.59 ± 0.25 C 3.25 ± 0.33 D 3.6 ± 0.31 C (range) 3.2-4.2 2.47-3.80 3.2-4.0 Cephalothorax width (mean) 2.75 ± 0.18 E 2.56 ± 0.24 F 2.75 ± 0.16 E (range) 2.4~3.2 2.04-3.10 2.6-3.0 Sample size 22 51 9 Females Total length (mean) 9.5 ± 0.7 G 8.09 ± 1.21 H 7.3-10.4 (range) 8.6-10.6 Cephalothorax length (mean) 4.0 ± 0.1 I 3.5 ± 0.4 J 4.0 ± 0.43 (range) 3.8-4.2 Cephalothorax width (mean) 3.0 ± 0.2 K 2.68 ± 0.35 L 3.02 ± 0.31 (range) 2.6-3.2 Sample size 8 61 7 of stridulatory organ at embolus base, scraper with two dark stripes down anterior side. Fang on distal tip of palpal tibia. furrow as in male. Femora I-IV yellow and Females: (Figs. 4, 5) {n = 8). Total length, annulated, patellae and tibiae light brown with cephalothorax length and width in Table 1 . Fe- annulations sometimes present on tibiae. Leg males slightly larger than males. Cephalotho- segment lengths similar to male lengths (Table rax brown, submarginal band narrow, distinct 2) except tibia and metatarsus shorter in fe- and wavy, pale median band as wide as pos- male. Abdominal dorsum with heart mark, ei- terior lateral eyes with slight indentation at ther distinct or faint (faint in 4 of 9 speci- posterior one third of band. Sternum brown to mens). and chevrons. Abdominal venter light brown with no spots, always darker than yellow coxae. Chelicerae brown, setaceous. yellow with black spots. Epigynum with ex- cavations on transverse piece (Fig. 4); exca- Table 2. — Comparison of length of segments of leg I in males {n — 21) and females (n = 8) of Schizocosa uetzi new species and males {n = 9) of Schizocosa rovneri. Measurements are in mm and means are ± SD. Schizocosa uetzi new species Schizocosa rovneri Females Males Males Femur 3.3 ± 0.2 3.56 ± 0.24 2.74 ± 0.96 Patella 1.6 ± 0.1 1.51 ± 0.12 1.36 ± 0.25 Tibia 2.7 ± 0.1 3.23 ± 0.23 2.63 ± 0.48 Metatarsus 2.6 ± 0.2 3.11 ± 0.75 2.59 ± 0.43 Tarsus 1.5 ± 0.1 1.66 ± 0.12 1.49 ± 0.26 Tibial width 0.5 ± 0 0.40 ± 0.1 0.40 ± 0.1 Ratio tibial length to width 5.4 8.1 6.6 88 THE JOURNAL OF ARACHNOLOGY Table 3. — Measurements of palps (n = 22) and epigyna (n - 8) of Schizocosa uetzi new species. Measurements are in mm and are given as means ± SD. Male Palp length 1.22 ± 0.33 Cymbium length 0.77 ± 0.15 Cymbium width 0.61 ± 0.06 Paleal process 0.42 ± 0.06 Female Total epigynal length 0.8 ± 0 Depth of hood 0.1 ± 0 Width, transverse piece widest part 0.7 ± 0.1 Width, longitudinal piece 0.2 ± 0.1 Height of excavation 0.1 ± 0 vations (EX) usually triangular in shape, slightly asymmetrical, nearly meeting at mid- line. Longitudinal piece sometimes narrowing anteriorly (4 of 9 specimens) or sides parallel (as in Fig. 4; 5 of 9) or narrowing slightly at midline. Spermathecae (SP) ovoid and smooth (Fig. 5). Courtship behavior. — ^The courtship be- havior of this species involves distinctive movements on the part of both the males and the females. The following descriptions are based on videotaped courtships of five court- ing pairs from Lafayette County, Mississippi; and three pairs from Natchitoches Parish, Louisiana. Males show chemoexploratory be- havior similar to that seen in other lycosids (Tietjen 1979) including S. ocreata, S. rov- neri, (Uetz & Denterlein 1979) and S. stri- dulans (Stratton 1991), wherein the male ex- Table 4. — Secondary sexual characteristics and courtship behavior, characteristics most useful in sepa- rating members of the Schizocosa ocreata species group that may be confused with Schizocosa uetzi new species (Dondale & Redner 1978; Uetz & Dondale 1979; Stratton 1991). Species Distinguishing features uetzi new species rovneri stridulans floridana crassipes ocreata Secondary sexual characteristics: Males have some pigmentation on tibia and sparse hairs on tibia; tibia and patella I always slightly darker than femur I. Courtship behavior: Pulses of stridulation. Secondary sexual characteristics: Lacking, mature males lack bristles or conspicuous pigmentation on legs 1. Tibiae I same color as femur of leg I. Courtship behavior: Body slams or “bounces” producing clear and distinct sounds (Uetz & Denterlein 1979). Secondary sexual characteristics: Males have black pigmentation on the tibia of legs I and halfway up the femur. Courtship behavior: Pulses of stridulation interspersed with tapping first pair of legs (Stratton 1991, in press). Secondary sexual characteristics: Legs I of mature male are slightly darker than other legs. Courtship behavior: Pulses that begin with two abdominal dips (producing a “squeak- ing” sound), followed by two pulses of stridulation, followed by two taps with the front legs. Other characters: Pale submarginal band of cephalothorax broken into three semicircu- lar patches that “break out” at carapace margins. Pale median band with undulating margins. Geographic distribution: limited to northern Florida and southern Georgia. Secondary sexual characteristics: Bristles on tibia of legs I black and dense. Courtship behavior: Behavior includes arch, extension and wave of legs I (Miller et al., in press). Secondary sexual characteristics: Bristles on tibia and metatarsus of legs I black and dense, bristles often extending to the basal portion of the tarsus. Courtship behavior: Active courtship involving extensive walking plus tapping and arching of legs 1. STRATTON— A NEW SPECIES OF SCHIZOCOSA 89 plores the substrate with the dorsal surface of his palp. Following chemoexploration, the male of S. uetzi new species typically per- forms several episodes of stridulation. In this behavior, the male assumes a stance with the body raised slightly from the substrate and stridulates by slight movements of the palps which are held nearly perpendicular to the substrate. The behavior is similar to that of S. stridulans, but there is no quick tapping of legs I as is seen with S. stridulans (Stratton 1991; Stratton in press). Females show a distinctive abdomen dip that occurs in the midst of male courtship. There was no sound recorded with the move- ment, and it was seen only in animals that eventually mated. Copulatory behaviors were very similar to the behaviors seen in S. ocreata, S. crassipes, S. rovneri, and S. stridulans. Males mounted so that the the male’s sternum was against the dorsal surface of the female’s abdomen. The male scraped his palp along the side of the female’s abdomen; she rotated her abdomen and his palp engaged her epigynum. There was a single expansion of the hematodocha, the palp disengaged, and the male then re-en- gaged the palp with another expansion of the hematodocha. After many engagements, he switched sides and repeated the sequence on the other side (Stratton et al. 1996). The du- rations for four copulations were 90 min, 115 min, 115 min, and 130 min. Geographic distribution, phenology and habitat. — Schizocosa uetzi new species has been collected from states throughout the mid- south region of the USA (Fig. 6). The species is the most common mid-sized wolf spider in June and July in northern Mississippi and northern Alabama; and it has been collected from Tennessee, western Arkansas and Loui- siana. A single individual, collected from South Carolina, had a maturation time consis- tent with this new species. Except for this sin- 90 THE JOURNAL OF ARACHNOLOGY (/) Schizocosa uetzi Schizocosa stridulans Schizocosa rovneri Figure 7. — Comparison of times of maturity of Schizocosa uetzi new species, Schizocosa stridulans and Schizocosa rovneri males in the southeastern USA, collections from 1991-1995. All individuals were caught by hand or pitfalls as adults. gle male, all populations are from west of the Appalachian Mountains. Extensive wolf spi- der collections have not yielded specimens of S. uetzi new species from Florida. Mature males have been collected from 7 June-2 August. There is broad overlap in phe- nology with S. stridulans (Fig. 7). Both S. rov- neri and the brush-legged species S. crassipes occur much earlier in the season (Figs. 7, 8). In a year-long pitfall study in Grenada County, Mississippi, there was almost no overlap be- tween S. uetzi new species and S. crassipes when the phenology of mature males was compared (Fig. 8). In Mississippi, Tennessee, Arkansas and Louisiana, S. uetzi new species has consis- tently been found in upland deciduous leaf lit- ter or upland deciduous litter mixed with pine litter. Schizocosa uetzi new species has fre- quently been collected with S. stridulans and shows broad geographic overlap with that spe- cies (Fig. 6, compare to fig. 14 in Stratton 1991) as well as overlap of habitat and phe- nology (Fig. 7). When S. uetzi new species and S. stridulans are collected together, S. uet- zi is slightly but significantly larger (as seen in Mississippi, Lafayette County, “Bailey’s Woods”, 15 June 1993; Student’s r-test, t = 5.91, 5.55, 6.48, P < 0.001 for cephalothorax width, length and tibial length). Additional material. — The following material was collected and identified as Schizocosa uetzi new species. ALABAMA: Jackson County: nr Russell Cave Natl. Monument, Id, day, 19 June 1992 (GES); Lauderdale County: Uplands of Ten- nessee River, West of Florence, 4d, 18 June 1984, (GES, LLW); Winston County: W.B. Bankhead Natl. Forest at Natural Bridge, Winston County Rd. #63 N of Houston, deciduous woods nr. ra- vine, 8d, night, 18 June 1992 (GES); Houston Campground, 2d, night, 18 June 1992 (GES). AR- KANSAS: Logan County: Mt. Magazine, Moss- back Ridge, South Slope, Id, pitfall, 23 June 1990 (B. Leary); Mossback Ridge, North Slope, 3d, pit- fall, 20 July 1990 (B. Leary). MISSISSIPPI: Claiborne County: Rocky Springs Park, Id, coll. 17 May 1983, matured in June (WPM); uplands woods about 10 mi S. of Vicksburg, 2d, coll. 20 May 1993, matured 14 June (GES, PRM). Gre- nada County: T21N R2E, Sec. 12, 13N, & R3E, Sec. 7S, 18N, 14 d, pitfall in deciduous woods, 5- 11 June 1991 (PRM, GES, GTB); 3d, day, 11 June 1991 (PRM, GES, TS, GTB); 15d, 19-25 June (GTB); 21 d, 26 June-2 July 1991 (PRM, GTB); pitfall on sandbar of creek, 12d, 26 June-2 July 1991 (PRM, GTB); deciduous woods, 5d, night, 26 June 1991 (PRM, KB); pitfall deciduous woods, lOd, 3-9 July 1991 (PRM); 2d, 10-16 STRATTON— A NEW SPECIES OF SCHIZOCOSA 91 Schizocosa crassipes Schizocosa uetzi Figure 8. — Comparison of occurrence of mature males of Schizocosa uetzi new species and mature males of Schizocosa crassipes from a pitfall study in Grenada County, Mississippi in 1991. July 1991 (PRM, GTB); IS, 17-23 July 1991 (PRM, GTB); IS, 26 July 1993 (GES, PRM, EAH, KW); 8S, coll. 21 May 1994, matured in lab 10, 13, 23 June 1994 (GES, PRM); T22N R3E, Sec. 31NW, pitfall deciduous woods, 12S, 19-25 June 1991 (PRM); IS, 10-16 July 1991 (GTB, PRM); deciduous woods by ravine. Id, 12 June 1991 (GES, PRM, TS, GTB). Lafayette County: Oxford, Id, 15 June 1984 (P.K. Lago); Old Taylor Rd., 2d, 11 June 1991 (PRM, GES); 4d, 29, night, 15 June 1991 (PRM, GES); deciduous woods, 3d, night, 15 June 1993 (PRM, GES, EAH); Puskus Lake, 13 mi NE Oxford, 8d, 11 June 1991 (PRM, WRM, GES); 3d, 21 July 1993 (GES, PRM, EAH, KW); 2 mi NW Oxford, decid- uous leaf litter, 1 d 1 9 , night, 30 June 1991 (PRM); Clear Creek Rec. Area, 2d, night, 17 June 1992 (GES); Bailey’s Woods, 12 d, night, 15 June 1993 (GES, PRM, GLM, EAH, WG, Young Scholars); Id, day, 15 June 1993 (EAH, KW); lOd, night (EAH, GLM, WG); 8 mi SE Oxford, TIOS R3W Sec.35; 34°36'N, 89°29'W, mixed pine and hard- wood, “Lonesome 80,” 19d, night, 10 June 1991 (GES, PRM, GLM, WRM, ME); Id, coll. 25 May 1992 (GES), spider sacrificed, 18 June 1992 (GES); rocky exposed hillside. Id, night, 15 June 1992 (GES); 2d, 1 July 1992 (PRM, GES); Id, night, 4 July 1992 (GES); 3d, 4 July 1992 (GES); day on hill by small lake, 4d, 23 June 1993 (GES, EAH, KW); 4d, night, 1 July 1993 (GES, EAH, PRM); pitfalls from 26 May, 1992 to July 1993, 9d, 3-10 June 1992 (GES, PRM); 5d, 16-24 June 1992, 6d, 24 June-1 July 1992, 17d, 1-8 July 1992, 12d, 8-15 July 1992, 14d, 15-22 July 1992, 2d, 22-29 July 1992, 4d, 10-20 June 1993, 2d, 20-28 June 1993, 14d, 28 June-5 July 1993, 12d, 5-12 July 1993, 14d, 12-21 July 1993 (GES, PRM). Marshall County: Wall Doxey State Park, T5S R3W Sect. 12, 89°24'W, 34°40'N, edge of deciduous woods, Id, coll. 23 May 1992, molt- ed 14 June (GES, PRM); nr. lake, 2d, 13 June 1991 (PRM, GES); nr. entrance to park, pine litter, 14 d, 1 pr. d & 9, night, 13 June 1991. Panola County: Sandstone Nature Trail nr Sardis Dam, in uplands on ridge, 13 d, night, 13 July 1993 (GES, PRM, EAH). Pontotoc County: 1 mi SE Ecru, pit- fall in deciduous woods, (4743-3,4), 2d, 5 June 1980 (PRM); Natchez Trace Parkway, Id in poor condition, kept in lab, 17 May 1983, 83-466 (WPM). Tishomingo County: Tishomingo St. Park, 3d, 21 June 1991 (GES, PRM). J.R Coleman State Park, oak pine woods along slope of a ravine, 2d 1 9 , 24 June 1986 (GES). TENNESSEE: Cum- berland County: Cumberland Mnt. State Park, de- ciduous woods, Id, night, 28 June 1983 (PRM). Dixon County: Montgomery Bell St. Park, up- lands, oak-pine woods. Id, day, 29 June 1992 (GES). Henderson County: Natchez Trace St. Park, Fair view Gully’s Trail nr, 1-40, oak-pine lit- ter, 3d, day, 29 June 1992 (GES). Marion County: Foster Falls Wild Area of S. Cumberland St. Rec. Area, 10 mi S. of Tracey City, 3d, 19 June 1992 (GES). Shelby County: Meeman Shelby State Park, uplands deciduous, at edge of woods, 3d, 15 July 1996 (E. Grey, D. Wells, GES, PRM). Wil- son County: Cedars of Lebanon State Park, Cedar Forest Loop Trail, hickory, oak, scattered cedar, 2d, 15 May 1993 (GLM). 92 THE JOURNAL OF ARACHNOLOGY KEY TO MATURE MALES IN THE SCHIZOCOSA OCREATA GROUP la. Males with thick brush of black bristles on tibia of legs I, sometimes extending to the basal region of the tarsus (thickness of bristles makes it difficult to see tibia); apparent width of tibia (with bristles) from lateral view more than twice width of tibia alone 2 lb. Males lacking thick brush of black bristles on tibia of legs I; may have some dark pigmentation or some dark hairs or may lack hairs and pigmentation on legs I ....... 3 2a. Paleal process with rugose prominence on retrolateral side (see Dondale & Redner 1978, Fig. 1; also Stratton 1991); tibial bristles extending to basal region of tarsus ocreata 2b. Paleal process with smooth prominence along its retrolateral side (see Dondale & Redner 1978, Fig. 2; also Stratton 1991); bristles on tibia only crassipes 3a. Males with dark pigmentation on tibia of legs I ....... 4 3b. Males lacking pigmentation on tibia and on femora. 5 4a. Males with dark pigmentation on tibia and on distal portion of femur ................ stridulans 4b. Males with pigmentation on tibia of legs I such that tibia is slightly darker than femur . . uetzi new species 5a. Pale median band on cephalothorax with edges parallel rovneri 5b. Pale median band on cephalothorax with edges scalloped ......................... floridana ACKNOWLEDGMENTS Financial support from the National Geo- graphic Society (grants #4916-92 and 5312- 94) to G.E. Stratton and G.L. Miller and Hew- let Mellon Faculty Research funds from Al- bion College to G.E. Stratton are appreciated. I thank G.L. Miller for use of videoequipment and access to lab and rearing chambers. I thank P.R. Miller for much help with field col- lections. In addition, I thank G.L. Miller, W.R. Miller, E.A. Hebets, E. Leighton, J. Hardy, J. Latimore, W. Garrison, K. White, G. Baker, T. Scheifer, W. Maddison, K. Benson, E. Grey, D. Wells and M. Eubanks for help with field collections. Some specimens were obtained with support from the William H. Cross Ex- pedition Fund of the Mississippi Entomolog- ical Museum and from NSF Grant BSR- 90244810 (R.L. Brown, RL). I also thank the officials of the Mississippi State Parks and Tennessee State Parks for permission to col- lect specimens. Thanks also to Marge Ramey for permission to collect on her land in Mis- sissippi. The drawings were done by J.C. Cok- endolpher. E.A. Hebets did some of the vid- eotaping. P.R. Miller assisted with making maps. The manuscript was improved by the close reading by J. Cokendolpher, C. Dondale, S. Reichling and C. Griswold. I am grateful for all of this help. LITERATURE CITED Dondale, C.D. & J.H. Redner. 1978. Revision of the Nearctic wolf spider genus Schizocosa (Ar- aneida: Lycosidae). Canadian EntomoL, 110: 143-181. Miller, G.L., G.E. Stratton, P.R. Miller & E.A. He- bets. In press. Geographic variation in male courtship behavior and sexual isolation in wolf spiders of the genus Schizocosa (Araneae; Ly- cosidae). Anim. Behav. Stratton, G.E. 1991. Schizocosa stridulans (Ara- neae: Lycosidae) a new species of wolf spider. J. ArachnoL, 18:29-39. Stratton, G.E., P.R. Miller, E.A. Hebets & G.L. Mil- ler. 1996. Pattern and duration of copulation in wolf spiders (Araneae, Lycosidae). J. ArachnoL, 24:186-200. Stratton, G.E. In press. Investigation of species di- vergence and reproductive isolation of Schizo- cosa stridulans (Araneae, Lycosidae) from Illi- nois. Bull. British ArachnoL Soc. Tietjen, W.J. 1979. Dragline following by male ly- cosid spiders. Psyche, 84:165-178. Uetz, G.W. & G. Denterlein. 1979. Courtship be- havior, habitat and reproductive isolation in Schi- zocosa rovneri Uetz & Dondale (Araneae: Ly- cosidae). J. ArachnoL, 7:121-128. Uetz, G.W. & C.D. Dondale. 1979. A new wolf spider in the genus Schizocosa (Araneae: Lycos- idae) from Illinois. J. ArachnoL, 7:86-87. Manuscript received 1 April 1996, revised 10 Sep- tember 1996. 1997. The Journal of Arachnology 25:93-96 RESEARCH NOTE THE EFFECT OF HABITAT STRUCTURE ON WEB HEIGHT PREFERENCE IN THREE SYMPATRIC WEB-BUILDING SPIDERS (ARANEAE, LINYPHHDAE) The quality of a foraging site can have a significant effect on the survival, growth and reproductive success of web-building spiders (Riechert & Tracy 1975; Lubin et al. 1993; Ward & Lubin 1993). Consequently, web-spi- ders can be expected to be found most often in areas where prey is abundant. Some spi- ders, such as Linyphiidae or Agelenidae, con- struct more or less permanent and costly webs (Janetos 1982); and a movement to another web site involves both the desertion of the old web and a high energy investment in web con- struction at the new site (Janetos 1986). Thus, web site selection is a particularly important issue for these spiders (Janetos 1982). Several factors may influence web site se- lection (Coleboum 1974; Uetz et al. 1978; Ol- ive 1980; Brown 1981; Pasquet 1984). For ex- ample, spiders may select web sites in order to exploit specific prey types (Cherrett 1964; Uetz et al. 1978; Olive 1980; Ward & Lubin 1993) or to utilize the physical characteristics of a web site (Robinson 1981; Greenstone 1984; Pasquet 1984; Bishop & Connoly 1992; Ehmann 1994). In the present study, web site selection in terms of web height was investigated for three sympatric linyphiid spiders, by testing wheth- er the presence of a surrounding understory vegetation can influence the web height se- lected on young conifer trees. The studied spi- ders, Frontinellina frutetorum (C.L. Koch 1834), Neriene radiata (Walckenaer 1841) and Linyphia triangularis Clerk 1757 con- struct three-dimensional sheet webs consisting of a centrally located platform with barrier threads above to intercept flying prey, knock- ing them to the platform where the spiders hang waiting underneath. Voucher specimens of each species were deposited in the Arach- noidea collection, at the Natural History Mu- seum Vienna, Austria. The study was conducted in a mixed decid- uous forest in eastern Austria, near Worth an der Lafnitz, approximately 15 km from Hart- berg (Styria). The study site (total area: 2854 m^) was comprised of plantations of Douglas fir {Pseudotsuga menziesii) and most webs were built on the young cultivated fir trees (Herberstein 1997). The study site was sub- divided into four plots and fenced in to protect the trees from browsing animals. The fencing allowed a dense understory of grasses, ferns, raspberry and blackberry bushes to grow around the trees, which was cut every fall as part of forestry management. An initial survey of web height (the dis- tance from the ground to the sheet of the web) in 1993 (Herberstein 1997) suggested that web height was not constant throughout the year but increased as the season progressed. This trend was confirmed in 1994. Ten tran- sects (10 X 1 m) were chosen each month (March-October) by randomly selecting the starting point and the direction (N, E, S, or W) of the transects which were allowed to in- tercept. Each inhabited web found along the transects was surveyed. There were significant positive correlations between web height (us- ing individual data points) and time of the year for F. frutetorum (r = 0.47, n = 152, P < 0.01), N. radiata (r — 0.48, n ~ 224, P < 0.01), and L. triangularis (r — 0.27, n = 287, P < 0.01) (Fig. 1). At the same time as the spiders’ web height increased, the vegetation surrounding the fir trees also increased in height, reaching its maximum height (Mean ± SD = 1.05 ± 0.33 m) in August/September. The observed change in web height may thus be a response to the growth of the understory, which over- grew web sites closer to the ground and re- duced their attractiveness. Consequently, webs on trees lacking a surrounding understory are 93 94 THE JOURNAL OF ARACHNOLOGY 3^ ^ 2.5-^ X 2-3 0 3 PQ 1 0.5 0 F. frutetorum "r I I 5 6 7 MONTH “1 10 3- ^ 2.5- ffi 2- O S1-5- ® 1. PQ ^ 0.5- L. triangularis I r 2 3 4 MONTH I ! I I I 8 9 10 Figure 1 . — The web heights are significantly cor- related with time of the season for Frontinellina frutetorum, Neriene radiata and Linyphia triangu- laris. expected to be placed closer to the ground compared to webs on trees surrounded by an understory. This assumption was tested during an experiment conducted from 10-15 August 1994 and from 2-13 September 1994. Twenty firs were selected in close proximity (within an area of 20 X 20 m), depending on the similarity of their height (mean ± SD = 2.03 ± 0.09 m), to reduce the effect of tree variability and habitat differences on web height selection. The trees were randomly aL located “control” or “experimental” trees. The vegetation surrounding the 10 experimen- tal trees was cut, leaving a clearance of 1 m in diameter, whereas the remaining 10 control trees were left in their natural condition, with a mature understory (maximum height: 1.2- 1.5 m) surrounding them. Before commencing the experiment, the 20 trees were surveyed twice (in the morning and the afternoon) and any spiders or webs on the trees were removed manually to avoid inter- ference by other spiders, previously present on the trees. As spiders may be attracted to areas where silk is present (which may bias the re- sults) the removal of spiders and web silk was carried out with great caution. Immature F. frutetorum and N. radiata and adult L. triangularis were collected in the morning and marked on the abdomen with red, nontoxic paint. Twelve hours later, each spider was released onto the lowest branch (0. 1-0.2 m from the ground) of each tree. Only a single spider was released per tree. The following morning the web height of each marked spider was measured, and the spiders and webs were removed. As not all spiders responded by construct- ing a web, the entire procedure was repeated five times. The data sets were distributed nor- mally (Kolmogorov-Smimov Goodness of Fit tests) and differences in web height on trees with and without an understory were analyzed using one-tailed r-tests. Bonferroni’s correc- tion (a' = a/k, where k equals the number of non-independent tests) was used to analyze the results (P = 0.05/3 = 0.017) in order to avoid inflation of the type I error probability. Removing the shrub layer had a significant effect on the position of the webs. The web heights of F. frutetorum (t = 2.5, df = 56, P = 0.0073), N. radiata (t - 3.05, df - 58, P — 0.0018) and L. triangularis {t = 2.5, df — 61, P = 0.0081) were significantly higher on trees surrounded by an intact understory than on trees without (Table 1). These results in- dicate that the vertical movement upwards may be a consequence of growing understory that either physically interferes with the spider webs, or reduces prey abundance to such an extent that the spiders desert their webs. Spiders may re-locate their webs in re- sponse to food supply (Olive 1982; Vollrath 1985; Gillespie & Caraco 1987), disturbance (Hodge 1987b), support structure (Enders 1974; Hodge 1987a; Bradley 1993), variation HERBERSTEIN— WEB HEIGHT PREFERENCE IN LINYPHIIDS 95 Table 1, — The average (mean ± SD) web heights of FrontinelUna frutetorum, Neriene radiata and Linyphia triangularis on trees with understory vegetation and without understory vegetation. Web heights (m) on trees Understory intact {n) Understory removed {n) FrontinelUna frutetorum 1.32 ± 0.30 (28) 1.10 ± 0.35 (30) Neriene radiata 0.79 ± 0.22 (27) 0.63 ± 0.21 (33) Linyphia triangularis 1.52 ± 0,40 (32) 1.27 ±0.38(31) in microclimatic conditions (Biere & Uetz 1981) or a combination of these factors. Spi- ders are unlikely to determine prey availabil- ity prior to web construction (Janetos 1986) or to use long term memory of site quality (Vollrath & Houston 1986). Thus, it seems un- likely that the web height selected by the spi- ders during the experiment is in direct re- sponse to prey abundance. Instead, the spiders might use microclimatic cues, which in turn may affect prey abundance. The results of this experiment demonstrate that web placement in spiders is selective and can be influenced not only by the actual sup- port structure utilized for web placement but also by the surrounding substrate such as a dense undergrowth. ACKNOWLEDGMENTS I thank Peter Dennis, Mark Elgar, Christian Kampichler, Norbert Milasowszky, Martin Predavec, Karl Sanger, Gerhard Spitzer, Klaus Peter Zulka and the reviewers of the manu- script for helpful comments on the manuscript and the study itself; Johann Anton Herberstein and Lisi Herberstein for their generous help; the University of Vienna for financial support. LITERATURE CITED Biere, J.M. & G.W. Uetz. 1981. Web orientation in the spider Micrathena gracilis (Walckenaer) (Araneae: Araneidae). Ecology, 62:336-344. Bishop, L. & S.R. Connoly. 1992. Web orientation, thermoregulation, and prey capture efficiency in a tropical forest spider. J. AractooL, 20:173-178. Bradley, R.A. 1993. The influence of prey avail- ability and habitat on activity patterns and abun- dance of Argiope keyserlingi (Araneae: Aranei- dae). J. Arachnol., 21:91-106. Brown, K.M. 1981. Foraging ecology and niche partitioning in orb-weaving spiders. Oecologia, 50:380-385. Cherrett, J.M. 1964. The distribution of spiders on the Moor House National Nature Reserve, West- morland. J. Anim. Ecol., 33:27-48. Coleboum, P.H. 1974. The influence of habitat structure on the distribution of Araneus diade- matus Clerk. J. Anim. Ecol., 43:401-409. Ehmann, WJ. 1994. Spider habitat selection: an experimental field test of the role of substrate diameter. J. Arachnol., 22:77-81, Enders, F. 1974. Vertical stratification in orb-web spiders (Araneidae, Araneae) and a consideration of other methods of coexistence. Ecology, 55: 317-328. Gillespie, R.G. & T. Caraco. 1987. Risk-sensitive foraging strategies of two spider populations. Ecology, 68:887-899. ^ Greenstone, M.H. 1984. Determinants of web spi- der species diversity: vegetation structural diver- sity vs. prey availability. Oecologia, 62:299-304. Herberstein, M.E. 1997, Niche partitioning in three sympatric web building spiders (Araneae: Liny- phiidae). Bull. British Arachnol. Soc. (In press). Hodge, M.A. 1987a. Microhabitat selection by the orb-weaving spider, Micrathena gracilis. Psyche, 94:347-361. Hodge, M.A. 1987b. Factors influencing web site residence time of the orb weaving spider, Mi- crathena gracilis. Psyche, 94:363-371. Janetos, A.C. 1982. Foraging tactics of two guilds of web-spinning spiders. Behav. Ecol. Sociobiol., 10:19-27. Janetos, A.C. 1986. Web-site selection: are we ask- ing the right questions? Pp. 9-22, In Spiders: Webs, Behavior, and Evolution. (WA. Shear, ed.). Stanford Univ. press, Stanford, California. Lubin, Y., S. Ellner & M. Kotzman. 1993. Web relocation and habitat selection in a desert widow spider. Ecology, 74:1915-1928. Olive, C.W. 1980. Foraging specializations in orb- weaving spiders. Ecology, 61:1133-1144. Olive, C.W. 1982. Behavioral response of a sit- and-wait predator to spatial variation in foraging gain. Ecology, 63:912-920. Pasquet, A. 1984. Predatory- site selection and ad- aptation of the trap in four species of orb-weav- ing spiders. Biol. Behav., 9:3-19. Riechert, S.E. & C.R, Tracy. 1975. Thermal bal- ance and prey availability: bases for a model re- lating web-site characteristics to spider reproduc- tive success. Ecology, 56:265-284. 96 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. Uetz, G.W., A.D. Johnson & D.W. Schemske. 1978. Web placement, web structure, and prey capture in orb- weaving spiders. Bull. British Ar- achnol. Soc., 4:141-148. Vollrath, F. 1985. Web spider’s dilemma: a risky move or site dependent growth. Oecologia, 68: 69-72. Vollrath, E & A. Houston, 1986. Previous experi- ence and site tenacity in the orb spider Nephila (Araneae, Araneidae). Oecologia, 70:305-308. Ward, D. & Y. Lubin. 1993. Habitat selection and the life history of a desert spider, Stegodyphus lineatus (Eresidae). J. Anim. EcoL, 62:353-363. Marie Elisabeth Herberstein’: Institute of Zoology, University of Vienna, Althanstr. 14, A- 1090 Vienna, Austria. Manuscript received 19 March 1996, revised 15 July 1996. ‘Current address: Department of Zoology, Uni- versity of Melbourne, Parkville Victoria 3052, Aus- tralia 1997. The Journal of Arachnology 25:97-98 RESEARCH NOTE ON SOME CAMILLINA FROM SOUTHERN AFRICA (ARANEAE, GNAPHOSIDAE) In Tucker’s (1923) survey of the ground spider fauna of southern Africa, 12 species were assigned to the genus Camillina Berland 1919. Of those, four have already been trans- ferred to other genera: Camillina acanthog- natha (Purcell 1907) to Trachyzelotes Loh- mander 1944 (by Platnick & Murphy 1984), C amnicola Tucker 1923 to Urozelotes Mel- lo-Leitao 1938 (by Platnick & Murphy 1984), and C browni Tucker 1923 and C lutea Tuck- er 1923 to Setaphis Simon 1893 (by Platnick & Murphy 1996). Of the others, C cordifera (Tullgren 1910), C procurva (Purcell 1908), and C. biplagia Tucker 1923 are currently considered valid species of Camillina (Plat- nick & Murphy 1987). Thus, five of the 12 species have not yet been treated in the modem literature. Through the courtesy of colleagues at the South Afri- can Museum in Cape Town, I’ve had the op- portunity to examine the recently rediscovered types of three of those species. One of these, Camillina postrema Tucker 1923, is represented by the male holotype from Diep River, Cape Flats, Cape Province, South Africa. It has the cheliceral bristles characteristic of Trachyzelotes and a palp characteristic of T. jaxartensis (Kroneberg 1875), a synanthropic and widespread species already recorded from South Africa. Like C. acanthognatha, C. postrema is here placed as a junior synonym of T. jaxartensis (NEW SYNONYMY). A second species, Camillina aestus Tucker Figures 1-4. — Camillina setosus Tucker. 1, Left male palp, ventral view; 2, Same, retrolateral view; 3, Epigynum, ventral view; 4, Same, dorsal view. 97 98 THE JOURNAL OF ARACHNOLOGY 1923, is represented by the female holotype from Nomptsas, Namibia. The epigynum is not that of a Camillina species, but bears a series of transverse ridges. Similar ridges oc- cur on the epigyna of two other species mis- placed by Tucker in Camillina: C. corrugata (Purcell 1907) and C. arida (Purcell 1907). Accurate placement of these three species must await study of their males; they could represent an aberrant species group of Zelotes Gistel 1848, or perhaps even of Urozelotes. The latter possibility is an interesting one, as it would offer the first real clues about the relationships and geographic origin of the widespread, synanthropic species U. rusticus (L. Koch 1872). A revision of the African spe- cies of Zelotes will be required to clarify the relationships of this species group. The third species, Camillina setosus Tucker 1923, is represented by one male and two fe- male syntypes from Signal Hill, Cape Town, Cape Province, South Africa. Platnick & Mur- phy (1987) indicated that this species was probably a true member of Camillina, but the types could not then be located, and no other specimens could be assigned to the name on the basis only of Tucker’s illustrations. Study of the now rediscovered syntypes indicates that this surmise was correct; C setosus is a valid member of Camillina, known only from the type specimens. As was suggested by Tucker, C setosus seems to be closest to C. biplagia; males share with that species a greatly elongated and sinuous embolus, but differ both in the shape of the embolus and of the terminal apophysis (Figs. 1, 2; cf. Platnick & Murphy 1987, figs. 37, 38). Females of C setosus can easily be distinguished from those of the other South African Camillina species by the widely separated posterolateral epigy- nal ducts (Figs. 3, 4). I thank C. Car, M. Cochrane, and H. Rob- ertson of the South African Museum for the loan of the types, and M.U. Shadab of the American Museum of Natural History for help with the illustrations. LITERATURE CITED Platnick, N.I. & J.A. Murphy. 1984. A revision of the spider genera Trachy zelotes and Urozelotes (Araneae, Gnaphosidae). American Mus. Novi- tates, 2792:1-30. Platnick, N.I. & J.A. Murphy. 1987. Studies on Malagasy spiders, 3. The zelotine Gnaphosidae (Araneae, Gnaphosoidea), with a review of the genus Camillina. American Mus. Novitates, 2874:1-33. Platnick, N.I. & J.A. Murphy. 1996. A review of the zelotine ground spider genus Setaphis (Ara- neae, Gnaphosidae). American Mus. Novitates, 3162:1-23. Tucker, R.W.E. 1923. The Drassidae of South Af- rica (Arachnida). Ann. South African Mus., 19: 251-438. Norman I. Platnick: Department of Ento- mology, American Museum of Natural His- tory, Central Park West at 79th Street, New York, New York 10024 USA. Manuscript received 13 February 1996, revised 1 September 1996. 1997. The Journal of Arachnology 25:99-105 RESEARCH NOTE A USEFUL PROCEDURE FOR ESTIMATING THE SPECIES RICHNESS OF SPIDERS Many authors have noted that the abiotic structure of the environment is particularly important to spiders (Luczak 1963; Lowrie 1973; Stratton et al. 1978; Uetz 1979; Hatley & McMahon 1980; Bultman & Uetz 1983; Gunnarson 1983, 1992; Greenstone 1984; Rushton 1991; Sundberg & Gunnarson 1994; Moring & Stewart 1995). By and large, spi- ders are without the biochemical mandates characteristic of many organisms; for exam- ple, they are not bound to particular plant spe- cies as are many insects. Spiders are often ex- ceptionally vagile and in addition may occupy different aspects of their environment as they mature. It is suggested here that given the dominant role of the structure of the habitat, a simple saturation model might best be used to estimate the species richness of a habitat. Some of the assumptions relevant to the de- velopment of the model include: 1) Rarely collected spiders are largely a consequence of the vagility of spiders and reflect to some de- gree the size of the regional pool, the propin- quity of other habitats, and the status of the populations of particular species at the time observations are made. For these reasons, spe- cies assemblages will tend to differ somewhat from year to year in any particular habitat (Rypstra & Carter 1995). 2) Spiders have fair- ly discreet requirements based on the spatial structure of the habitat and to a lesser degree on other factors such as humidity, tempera- ture, light intensity, etc. (Rushton 1991; Mor- ley & Stewart 1995). 3) Some species of spi- ders have different spatial requirements as they mature. 4) Within the regional pool there is a species assemblage that is adapted to a considerable degree to the niche-spatial op- tions at any particular time offered by a hab- itat (more exactly, perhaps, the contained set of “microhabitats”). Spiders that are not suit- ed to a particular habitat and wander in may soon leave or become prey for other organ- isms, including other spiders that are well adapted to the habitat. 5) Combining samples taken in different years may lead to an over- estimate of the number of species character- istic of any particular habitat as a result of the accumulation of records of species that result simply from the vagaries of vagility (Edwards 1997). As the model developed, it soon be- came apparent that it was analogous to the equation for adsorption isotherms created by Langmuir (1918). It is postulated that within any habitat there is a set of n species or species-specific niches representing the maximum potential number of species in the habitat. At any time, of these niches are occupied. When the habitat is saturated, = n. The number of n potential and occupied niches may vary with season, with the availability of other suitable habitats, and with changes in the regional pool (im- migrants, introduced species, population changes both within and without the spider community that modify interactions, etc.). Let n ~ total number of species-specific niches. The rate of entry into the habitat: dnjdt - k„(n-nq)q where ko. = rate constant of arrival in the hab- itat and q = number of samples (quadrats). The rate of species entrance is proportional to the number of unoccupied niches, and to the sampling intensity. The rate of species disappearance is: -dnjdt -= where — the rate constant of species dis- appearance. Rate of species disappearance is proportional to space already occupied. De- parture may be voluntary, but includes other factors such as predation and disease. Equating entry and departure (equilibrium): kjn-njq = k^n^ (1) Taking reciprocals after setting equation equal to q\ 99 100 THE JOURNAL OF ARACHNOLOGY 1 / AREA SAMPLED (M SQ) NO. SPECIES J(ESR) 1 / AREA SAMPLED (M SQ) NO. SPECIES J(ESR) 1 / AREA SAMPLED Figure 1. — Six examples of Cape Cod habitat data, with the reciprocals of the number of quadrats plotted against the reciprocals of the number of species collected and the reciprocals of the number of species calculated using the jackknife estimator. Each of these was sampled over a SVz month period, 15 June-September 1989 and 1990. 1/q = kXn-n^)fk^n^ Rearrangement gives: 1/n^ = [(k,/k,n)(l/q)] +l/n (2) Note that a plot of 1/nq against I/q is a straight line, with the slope = k^/k^n and the intercept = 1/n. The reciprocal of the intercept provides the estimated number of species, n(esr), at saturation, and n = n^. In Figs. 1 and 2 both the number of species at saturation, n( esr) and the calculated number of species using the jackknife estimator, developed by Heltsche & Forrester (1983) are shown. The data were resampled randomly EDWARDS— ESTIMATING SPECIES RICHNESS 101 Figure 2. — Plots of the reciprocals of the number of quadrats against the reciprocals of the number of species and number of species calculated using the jackknife estimator for two examples from Costa Rica and four from El Junque, Puerto Rico. The Costa Rican samples were each collected in one day, while those from Puerto Rico were collected at intervals during 1994 and 1995. without replacement, at three quadrat intervals beginning at 10 quadrats, with 100 iterations for each level of quadrat aggregation. The graphs are double reciprocal plots, with the number of quadrats shown on the abscissa and the number of species on the ordinate. In Fig. 1, plots of the results of analysis for six habitats from the Cape Cod region are pre- sented. These are some of the habitats and col- lections reported on earlier (Edwards 1993). These habitats were sampled from the middle of June to the end of September, 1989-1990. In Fig. 2 data for various habitats in Puerto Rico and Costa Rica are plotted. The Costa 102 THE JOURNAL OF ARACHNOLOGY Table 1. — Statistical data for habitats using linear regression Mn^ = {x){yq) + c (Equation 5). q = number of quadrats, c = constant, SE = standard error, x = slope, n{esr) = estimated total number of species (reciprocal of c), for calculations based on species, q^Q = number of quadrats required to collect 50% of n{esr),j{esr) and for calculations based on number of species derived from jackknife estimator. Data in descending order of estimated total number of species n{esr). For habitat codes see Table 2. Code c SE c X SEx n(esr) ^50 j(esr) CL 50 0.0062 0.0001 0.0899 0.0011 162.09 16.2 0.998 197.23 0.999 DL 41 0.0069 0.0001 0.0911 0.0013 144.91 14.1 0.998 177.38 0.998 CP 123 0.0077 0.0004 0.2122 0.0040 129.73 27.5 0.994 179.92 0.992 DP 41 0.0088 0.0003 0.3011 0.0039 113.25 34.1 0.999 160.23 0.999 XL 80 0.0093 0.0003 0.1277 0.0027 108.07 13.7 0.990 131.35 0.993 GP 52 0.0097 0.0005 0.2724 0.0055 103.07 29.0 0.995 140.31 0.992 DU 45 0.0098 0.0002 0.1037 0.0020 101.62 10.4 0.996 125.33 0.990 FP 43 0.0100 0.0002 0.2204 0.0022 100.16 21.9 0.999 123.82 0.996 CU 47 0.0100 0.0002 0.1560 0.0024 99.65 15.6 0.998 111.75 0.995 RU 23 0.0101 0.0002 0.2063 0.0007 99.31 20.5 1.000 158.73 0.997 FS 53 0.0112 0.0004 0.1198 0.0041 89.63 10.7 0.985 117.47 0.960 GS 45 0.0113 0.0006 0.1899 0.0075 88.18 17.0 0.985 130.04 0.984 PF 40 0.0115 0.0002 0.1422 0.0032 86.98 12.4 0.995 111.35 0.996 JF 44 0.0125 0.0003 0.1229 0.0035 79.76 9.8 0.992 99.59 0.979 DT 41 0.0149 0.0006 0.2856 0.0085 67.10 19.0 0.992 101.05 0.990 GU 40 0.0163 0.0005 0.2582 0.0069 61.37 15.9 0.994 78.65 0.988 CT 40 0.0165 0.0005 0.2575 0.0069 60.68 15.6 0.994 87.21 0.990 SF 25 0.0200 0.0002 0.1985 0.0031 50.08 10.4 0.998 61.82 0.996 HL 32 0.0206 0.0004 0.2224 0.0058 48.54 10.8 0.996 60.36 0.994 MU 71 0.0231 0.0010 0.2793 0.0100 43.35 12.1 0.976 56.38 0.967 ML 46 0.0244 0.0005 0.3688 0.0054 40.93 15.1 0.998 49.34 0.989 HU 39 0.0250 0.0010 0.2595 0.0143 39.97 10.4 0.976 51.38 0.967 BL 30 0.0256 0.0007 0.3337 0.0120 39.09 13.1 0.994 53.71 0.991 TL 61 0.0266 0.0005 0.3953 0.0048 37.53 14.8 0.998 41.99 0.967 BU 32 0.0271 0.0006 0.3408 0.0096 36.96 16.3 0.995 51.26 0.989 FU 14 0.0338 0.0019 0.1451 0.0090 29.63 4.3 0.992 45.52 0.996 VU 22 0.0355 0.0005 0.2790 0.0115 28.16 7.9 0.995 37.11 0.987 Means 15.5 0.993 0.988 Rican habitats were each sampled in one day. The Puerto Rican habitats show data collected at periodic intervals during 1994 and 1995. The mean values (Table 1) for the esti- mated number of species, n(esr), against num- ber of quadrats was == 0.9933 (0.9761- 0.9999), while the mean value of for cal- culations based on the jackknife estimator, j(esr), was - 0.9884 (0.9597-0.9985). The slightly increased variability of the jackknife data is apparent to the eye in the figures. Es- timates of j(esr) averaged about 31% more than n(esr). In some habitats there is an increase in the slope of the fitted line toward the origin, as may be seen in Fig. 1 for the red cedar foliage (IF) and in East Falmouth mixed forest leaf litter (XL). In Fig. 3, the skew (gj) and kur- tosis (^2) of the frequency distribution of the numbers of species/quadrat in each habitat for the Cape Cod area is portrayed. The species assemblages of spiders are typically positively skewed and leptokurtic. The downward bend exhibited in some plots is evidence of the ex- istence of a platykurtic and/or a multimodal distribution. The sampling period for the Cape Cod hab- itats included at least part of the early fall on- set of a different assemblage of species. The frequency distribution of the number of spe- cies for red cedar foliage (IF) is platykurtic and distinctly bimodal (Fig. 4a). Separating the data for the months of June- July from that for August-September resulted in the fitted lines shown in Fig. 4b. The second mode is interpreted as representing the incoming as- EDWARDS— ESTIMATING SPECIES RICHNESS 103 0.2 -1 CAPE COD HABITATS ■ GP ■ CU ■ DP CP *■ SF ■ FP XL ■ PF* *■ DT JF ■ CL ■ CT ■ DU ■ GS FS ■■ DL 5 -1 -0.5 0 0.5 1 1.5 2 KURTOSIS Figure 3. — The central moments, skew (gl) and kurtosis (g2) for the frequency distribution of the number of species/quadrat in Cape Cod habitats. Habitat codes are given in Table 2. A platykurtic distribution often indicates temporal change and/or environmental disturbance during the period of sampling. This is indicated by the increasing slopes in the red cedar foliage and mixed forest leaf litter data shown in Fig. 1. semblage, with the vanishing summer assem- blage contributing largely to the first mode. The n(esr) for this entire data set was 79.8 species (r^ = 0.992). The n(esr) for June and July was 62.3 (r^ == 0.9999) and for August and September was 74.7 (r^ = 0.9971). Sim- ilarly the pine foliage (PF) and deciduous trunk (DT) samples also suggested that sea- sonal change was involved. The East Fal- mouth mixed forest leaf litter (XL) data was more difficult to interpret. This habitat was sampled from January-March in 1993 (see Fig. 4c). By and large, the collection con- tained species to be expected in litter during the colder months. Also present was a fairly large number of arboreal species that could be considered the constituents of a second dif- ferent assemblage, some or all individuals of which had taken refuge in the litter during the winter months. Temporal change in the species assem- blages during the period of sampling is a pos- sibility that must be recognized. It should not, however, mitigate against comparable sam- pling from one year to the next providing rel- evant factors are kept constant. Two examples of habitats sampled in Costa Rica are shown in Fig. 2. These habitats were each sampled in one day. The Puerto Rico samples were taken in the rainforest on El Junque during periodic visits from May 1994- NO. SPECIES -«•- J(ESR) Figure 4. — 4a, Frequency distribution of the number of species taken in red cedar (juniper) fo- liage (RF); 4b, Plot of the number of species for the period June-July, and the period August-Sep- tember; 4c, The frequency distribution of species in the East Falmouth mixed forest leaf litter habitat (XL), collected January-March 1993. March 1995 in the understory of a mahogany plantation (ML, MU) and September 1994- March 1995 in the leaf litter at Palo Hueco, a young mixed second forest area (HL, HU). During this period the area suffered a severe drought. This event may have contributed to some of the irregularity shown in the under- story samples, although not in the leaf litter collections. An examination of the mahogany 104 THE JOURNAL OF ARACHNOLOGY Table 2. — List of codes used in Table 1, locality sampled, habitat, and estimated quadrat area and sampling method. CC = Cape Cod, CR = Costa Rica, PR = Puerto Rico (El Junque). For further details on sampling methods, see Edwards (1993). Code Locality Estimated quadrat area CL Coniferous leaf litter, CC. 0.25 m^ — sample DL Deciduous leaf litter, CC. 0.25 m^ — sample CP Coniferous forest pitfall, CC. ±3.0 m^ — sample DP Deciduous forest pitfall, CC. ±3.0 m- — sample XL Mixed forest leaf litter, CC. 0.25 m^ — sample GP Grass field pitfall, CC. ±3.0 m^ — sample DU Deciduous forest understory, CC. 2.4 m^ — sweeping FP Old field pitfall, CC. ±3.0 m- — sample cu Coniferous understory, CC. 3.0 m^ — beating RU Guapiles rainforest understory, CR. 3.0 m’ — sweeping FS Old field foliage, CC. 10.0 m^ — sweeping GS Grass field foliage, CC. 10.0 m“ — sweeping PF Pine foliage, CC. 1.0 m^ — beating IF Red Cedar foliage, CC. 1.0 m- — beating DT Oak trunk, CC. 1.0 m^ — brushing GU Taboga Gallery Forest understory, CR. 10.0 m- — sweeping CT Pitch pine trunk, CC. 1.0 m- — bark removal SF Spruce foliage, CC. 1.0 m^ — beating HL Palo Hueco mixed forest litter, PR. 0.25 m^ — sample MU Mahogany forest understory, PR. 3.0 m- — sweeping ML Mahogany forest leaf litter, PR. 0.25 m^ — sample HU Palo Hueco understory, PR. 3.0 m“ — sweeping BL Mt. Britten Dwarf Forest leaf litter, PR. 0.25 m- — sample TL Tabonuco forest leaf litter, PR. 0.25 m- — sample BU Mt. Britten Dwarf Forest understory, PR. 3.0 m- — sweeping FU Guapiles rainforest fern understory, CR. 10.0 m^ — sweeping VU Monteverde cloud forest understory, CR. 3.0 m‘ — sweeping understory data showed some ambiguous ev- idence of seasonal change. There is one further attribute of the satu- ration model that should be noted. At equilib- rium (Equation 1), the number of quadrats needed under ideal circumstances to achieve 50% of the estimated number of species may be calculated. This is simply the reciprocal of the intercept/slope of Equation 2 (see Table 1 , q^o). In general, with the exception of the pit- fall trap collections, when the number of quadrats reaches ±16, 50% of the estimated total number of species has been achieved (Table 1). In the case of pitfall traps, the num- ber of quadrats needed to reach 50% is con- siderably more. Again, with the exception of pitfall trap collections, approximately 40 quadrats were necessary to achieve ±75% of the estimated total number of species. The ground level is the principal route traveled by wandering spiders, thus pitfall traps provide a sample of the immediate habitat as well as a sample of wandering spiders, especially adult males, from other habitats. The number of quadrats required to reach the q^Q level (or any other desired level) is not directly determined by the number of species to be found in the habitat, or the number of quadrats, but rather by the rate at which additional species enter the collection. Within reasonable limits the size of the quadrat chosen should not alter the n(esr) although smaller quadrats are intuitive- ly to be preferred, particularly where there is evidence of aggregation. The sampling meth- od used and an estimate of the area sampled for the data presented here is provided in Ta- ble 2. In summary, the data appear to be well fit- ted using a double reciprocal plot of the num- ber of quadrats against the number of species. Simply using the number of species observed results in a relatively easy to understand es- timate of species richness since there are no abstract mathematical considerations in- EDWARDS— ESTIMATING SPECIES RICHNESS 105 volved. The procedure appears to be amenable to interpretation; that is, identifying and quan- tifying temporal change and other environ- mental changes. It is suggested that the jack- knife estimator of Heltshe & Forrester (1983), an approach that involves giving weight to the number of unique species, will tend to over- estimate the number of species. Combining samples taken in different years will result in an ever increasing overestimate of the species richness of a particular habitat, given the va- gility of spiders. The regional pool of spiders in the Cape Cod, Massachusetts area is estimated to be ap- proximately 500 species (Edwards 1993). The results presented (see species estimates in Ta- ble 1) support the suggestion that specific Cape Cod habitats have a high beta diversity, are inhabited by interactive assemblages of spiders and tend to be saturated (Cornell & Lawton 1992; Edwards 1997). ACKNOWLEDGMENTS The efforts of Eric Edwards collecting spi- ders over the years is very much appreciated. John Edwards, Dept, of Chemistry, Brown University, assisted in the development of the proposed model. I wish to thank Herbert Nan- ne (Ministry of Agriculture and Animal Hus- bandry, Costa Rica), and Gerardo Camilo (University of Puerto Rico, Puerto Rico) for various kindnesses that expedited work in the field. Conversations with and encouragement from Jonathan Coddington (United States Na- tional Museum) over the years have been more than helpful. The comments and con- structive criticisms of several anonymous re- views were helpful and gratefully received. Wendy L. Gabriel’s (Northeast Fischeries Center, Woods Hole) helpful critique of the revised manuscript was very much appreciat- ed. LITERATURE CITED Bultman, TL. & G.W. Uetz. 1983. Abundance and community structure of forest floor spiders fol- lowing litter manipulation. Oecologia, 55:34-41. Cornell, H.V. & J.H. Lawton. 1992. Species inter- actions, local and regional processes, and limits to the richness of ecological communities; a the- oretical perspective. J. Anim. EcoL, 61:1-12. Edwards, R.L. 1993. Can the species richness of spiders be determined? Psyche, 100:185-208. Edwards, R.L. 1997. Behavior and niche selection by mailbox spiders. J. ArachnoL, 25:00-00. Greenstone, M.H, 1984. Determinants of web spi- der species diversity; vegetation structural diver- sity vs. prey availability. Oecologia, 62:299-304. Gunnarsson, R. 1983. Winter mortality of spruce living spiders; effects of spider interactions and bird predation. Oikos, 40:226-233. Gunnarsson, R. 1992. Fractal dimension of plants and body size distribution of spiders. Funct. EcoL, 6:636-641. Heltsche, J.F. & N.E. Forrester. 1983. Estimating species richness using the jackknife procedure. Biometrics, 39:1-11. Hatley, C.L. & J.A. MacMahon. 1980. Spider com- munity organization; seasonal variation and the role of vegetation architecture. Environ. Ento- moL, 9:632-639. Langmuir, I. 1918. The adsorption of gases on plane surfaces of glass, mica and platinum. J. American Chem. Soc., 40:1361-1403. Lowrie, D.C. 1973. The microhabitats of western wolf spiders of the genus Pardosa. Ent. News, 84:103-116. Luczak, J. 1963. Differences in the structure of communities in web spiders in one type of en- vironment (young pine trees). EcoL Polska Acad., 11:159-221. Moring, J.B. & K.W Stewart. 1994. Habitat par- titioning by the wolf spider (Araneae, Lycosidae) guild in streamside and riparian vegetation zones of the Conejos River, Colorado. J. ArachnoL, 22: 205-217. Rushton, S.P. 1991. A discriminant analysis and logistic regression approach to the analysis of Walckenaeria habitat characteristics in grassland (Araneae: Linyphiidae). Bull. British ArachnoL Soc., 8:201-208. Rypstra, A.L. & PE. Carter. 1995. The web-spider community of soybean agroecosystems in South- western Ohio. J. ArachnoL, 23:135-144. Stratton, G., G. Uetz & D. Dillery. 1978. A com- parison of the spiders of three coniferous tree species. J, ArachnoL, 6:219-226. Sundberg, I. & R. Gunnarsson. 1994. Spider abun- dance in relation to needle density in spruce. J. ArachnoL, 22:190-194. Uetz, G.W. 1979. The influence of variation in lit- ter habitats on spider communities. Oecologia (Berlin), 40:29-42. Robert L. Edwards: Box 505, Woods Hole, Massachusetts 02543-0505 USA. Manuscript received 31 August 1995, revised 15 July 1996. 1997. The Journal of Arachnology 25:106-108 RESEARCH NOTE A NOTE ON EUSCORPIUS CARPATHICUS (SCORPIONES, CHACTIDAE) FROM THE CRIMEA Euscorpius carpathicus (L. 1767) (Chacti- dae), a scorpion species fairly common in southern Europe where it ranges from Spain to Ukraine, has been extensively studied (e.g., Birula 1917; Hadzi 1930; Caporiacco 1950; Vachon, 1963, 1975, 1978; Curcic 1972; Kin- zelbach 1975; Fet 1986; Sherabon, 1987). There are 24 described subspecies; and, for most, the taxonomic status is unclear. Many of these forms are somewhat geographically isolated; for example, nearly every Mediter- ranean island (e.g., Mallorca, Sardinia, Sicily, Crete) has an endemic subspecies. The Crimea Peninsula (currently an admin- istrative territory within Ukraine) houses the easternmost, disjunct population of E. car- pathicus. It is the only species of scorpion found in the Crimea. This population was first recorded from Alupka by Pallas (1795). It was described by C.L. Koch (1838) as Scorpius tauricus and for many years was treated as a separate, endemic species. Birula (1917) listed it as Euscorpius tauricus (C.L. Koch) and gave a detailed description of its anatomy and biology. Caporiacco (1950) synonymized it as a subspecies of Euscorpius carpathicus (L.). The original material from the Crimea has not been analyzed since 1917. The studied sample included 71 specimens (17(3, 54$) from the following localities of the Crimea Peninsula (area between 33-35°E and 44-45°N): Alushta, Balaklava, Frunzen- skoye, Gaspra, Inkerman, Kerch’, Nikitsky Botanical Garden, Oreanda, Sevastopol’, Sim- eiz, Simferopol’, Sudak, Yalta, Yevpatoria. The studied specimens are deposited in the Zoological Institute of the Russian Academy of Sciences (St. Petersburg, Russia) and in the Zoological Museum of the Moscow State Uni- versity (Moscow, Russia). Detailed label data are published in Fet (1989). The majority of this material originated from the Black Sea coast (southern parts of the peninsula), known for its mild climate due to the protection of the Yaila range which runs latitudinally across the peninsula. Following the technique developed for scorpions by Vachon (1963, 1975), I scored numbers of trichobothria on the pedipalp pa- tella, which, in Euscorpius, vary both among and within local populations. Ventral tricho- bothria form a single row (Tv), whereas ex- ternal ones appear in six clusters: terminal (er), subterminal {est), median (em), suprabas- al {esb), and two basal groups {eb„ and eb). There is no sexual dimorphism. Numbers may vary between left and right pedipalp, but such asymmetry is a subject of a separate study. Trichobothrial numbers scored for the Cri- mean population were: Tv = 7 (20 cases, 14.3%), Tv = 8 (119 cases, 85.0%) and Tv = 9 (1 case, 0.7%%) (number of scored pedi- palps, 140); et = 5 {\A cases, 10.0%), et = 6 (126 cases, 89.4%) and et = 7 (I case, 0.6%) {n = 141). Numbers of external trichobothria in other five groups did not vary and were: est = 4, em = 4, esb = 2, eb^ = 4, eb = 4. Although some authors (Curcic 1972; Kinzel- bach 1975) attempted to discuss clinal varia- tion in trichobothria within E. carpathicus, few data are published that can be used for comparison to the population above. Kinzel- bach (1975) gave an qualitative overview of many samples from the Balkan Peninsula and the Aegean Sea islands, using only the Tv in- dex. He recognized not one but two species: an “oligotrichous” E. carpathicus (L.) with Tv = 7-8 and a “polytrichous” E. mesotri- chus Hadzi with Tv = 10-12, which produce hybrid “mesotrichous” forms with Tv — 9- 10. This division was not accepted by other authors (Vachon 1978; Fet 1986, 1989; Scher- abon 1987). The Crimean population has values of Tv close to 8 (mean Tv = 7.86, s^ = 0.13) and et close to 6 (mean et = 5.90; s^ = 0.10). Tv from 7-8 and et = 6 are found in certain pop- ulations from northeastern Greece (Kinzel- 106 FET—EUSCORPIUS FROM THE CRIMEA 107 bach 1975; Fet 1986), On the other hand, tri- chobothrial numbers of Tv = 9-10 and et = 7, which are common throughout the Balkans and Crete (Fet 1986), are very rare (< 1%) in the analyzed Crimean sample. Populations of E. carpathicus farther westward are charac- terized by the forms with higher values of Tv = 10“12, and et ^ 7-8, e.g., in Austria (mean Tv = 10.25, s^ = 1.04; mean et " 7.47, s^ = 0.38; Scherabon 1987) or Sardinia (mean Tv = 1 1.01, s^ = 0.36; mean et = 7.32, s^ = 0.67; Vachon 1978). Means of trichobothrial scores of the Austrian and Sardinian populations are not significantly different; r-values are 0.93 for Tv {P > 0.5) and 0.28 for et {P > 0.7). However, the mean of the Crimean population significantly differs from a combined Austria/ Sardinia sample (mean Tv = 10.67, s^ = 0.30; mean et = 7.40; s^ = 0.40). For this compar= ison, ^-values are 6.85 for Tv (P < 0.001) and 2.74 for et (P < 0.01). According to Kinzel- bach’s (1975) terminology, the Crimean pop- ulation is the sensu stricto “oligotrichous” E. carpathicus (L.). The isolated zoogeographic position of this Crimean scorpion, and that of many Crimean animal and plant populations, is unique for the species’ range: the closest populations of E. carpathicus are those in Romania, about 500 km westward. The reason for such disjunction should be sought in the paleogeographical his- tory of the Crimea, which is relatively well studied (Golovach 1984). This area originated as an island of the Tethys Sea during the Me- sozoic and throughout the Tertiary period was connected many times to different land mass- es (Caucasus, Balkan Peninsula, Anatolia, an- d/or modem Ukraine) when the sea regressed. There are no Tertiary relicts in the Crimea; and all endemic plants there are generally very recent (Grosset 1979). Severe Pleisto- cene glaciations in Europe (the last one, the Wurm Ice Age, 70,000-11,000 years BP, cor- responds to the Wisconsin of North America) could have eliminated most of ancient ther- mophile and mesophile Mediterranean species of the Crimea. Golovach (1984) analyzed the diplopod fauna in the Crimea, and suggested that its age is primarily Pleistocene and the source of migration was the eastern Mediter- ranean, especially the Balkan Peninsula. It can be suggested that the existence of E. c. taur- icus is the a result of a (possibly recent) mi- gration from the Balkan Peninsula in the Pleistocene interglacials. The source of such migration, then, should have been ‘'oligotri- chous” populations of eastern Balkans (with Tv = 7-8 and et = 6). Further comparative studies should assess the criteria for subspe- cific stmcture of £*. carpathicus. ACKNOWLEDGMENTS I thank Vladimir 1. Ovtsharenko (St. Pe- tersburg) and Kirill G. Mikhailov (Moscow) for their permission to work with Russian mu- seum collections; and W. Starega (Warsaw, Poland) for the loan of material. I am es- pecially thankful to Mikhail Eidelberg, Sergei Sharygin, and Konstantin Yefetov who col- lected scorpions from the Crimea on my re- quest, and to Matt Braunwalder (Zurich, Swit- zerland) for his constant and invaluable help in obtaining literature. LITERATURE CITED Birula, A. (Byalynitsky-Birala, A. A.). 1917. Fauna of Russia and adjacent countries. Arachnoidea. VoL I. Scorpions. Petrograd, 224 pp. (in Rus- sian); English translation by Israel Program for Scientific Translations, Jerusalem 1965, 154 pp. Caporiacco, L. di. 1950. Le specie e sottospecie del genre "'Euscorpius'^ viventi in Italia ed in alcune zone confinanti. Memorie/Accademia na- zionale dei Lincei. Classe di scienze fisiche, ma- tematiche e naturali, (8) 2 Scz. 3a, 4:159-230. Curcic, B.P.M. 1972. Considerations upon the geo- graphic distribution and origin of some popula- tions in the genus Euscorpius Thorell (Chactidae, Scorpiones). Rapp. Comm. Intemat. Mer Medi- terranee, Monaco, 21:83-88. Fet, V. 1986. Notes on some Euscorpius (Scorpi- ones, Chactidae) from Greece and Turkey. Riv. Mus. Scien. Natur. E. Caffi (Bergamo), 9:3-11. Fet, V. 1989. A catalogue of scorpions (Chelicer- ata: Scorpiones) of the USSR. Riv. Mus. Scien. Natur. E. Caffi (Bergamo), 13:73-171. Golovach, S.L 1984. Distribution and faunogenesis of the Diplopoda of the European USSR. Pp. 92- 138, In Faunogenesis and Phylocenogenesis. (Chernov, Yu.L, ed.). Nauka, Moscow (in Rus- sian). Grosset, G.E. 1979. On the origin of flora of the Crimea. Byulleten MOIP (Bull. Moscow Soc. Natur.), Div. Biol., Part 1; 84:64-84; Part 2; 84: 35-55 (in Russian). Hadzi, J. 1930. Die europaischen Skorpione des Polnischen Zoologischen Staatsmuseums in War- szawa. Annal. Mus. Zool. Polonici, 9:29-38. Kinzelbach, R. 1975. Die Skorpione der Agais. Beitrage zur Systematik, Phylogenie und Biogeo- graphie. Zool. Jahrb. (Syst.), 102:12-50. 108 THE JOURNAL OF ARACHNOLOGY Koch, C.L. 1838. Die Arachniden. C.H. Zeh’sche Buchhandlung, Niimberg, 4:1-144. Pallas, P.S. 1795. A brief physical and topograph- ical description of the Taurian Region. St. Pe- tersburg (in Russian). Scherabon, B. 1987. Die Skorpione Osterreiches in vergleichender Sicht unter besonderer Beriicksi- chtung Kamtens. Pp. 77-154, In Karinthia II, 45. Sonderheft, Klagenfurt. Vachon, M. 1963. Remarques sur 1’ utilisation, en systematique, des soies sensorielles (trichoboth- ries) chez les Scorpions du genre Euscorpius Thorell (Chactidae). Bull. Mus. Nat. d’Hist. Na- tur. (Paris), 34:347-354. Vachon, M. 1975. Recherches sur les Scorpions appartenant ou deposes au Museum d’Histoire naturelle de Geneve. 1. Rev. Suisse Zook, 82: 629-645. Vachon, M. 1978. Remarques sur Euscorpius car- pathicus (Linne, 1767) canestrinii (Fanzago, 1872) (Scorpionida, Chactidae). Annal. Hist.-Na- tur. Mus. Nat. Hungarici, 70:321-330. Victor Fet: Department of Biological Sci- ences, Marshall University, Huntington, West Virginia 25755-2510 USA Received 1 March 1996, revised 11 October 1996. INSTRUCTIONS TO AUTHORS (revised October 1996) Manuscripts are preferred in English but may be ac- cepted in Spanish, French or Portuguese subject to availability of appropriate reviewers. Authors whose pri- mary language is not English may consult the Associate Editor for assistance in obtaining help with English manuscript preparation. All manuscripts should be pre- pared in general accordance with the current edition of the Council of Biological Editors Style Manual unless instructed otherwise below. Authors are advised to con- sult a recent issue of the Journal of Arachnology for additional points of style. Manuscripts longer than 1500 words should be prepared as Feature Articles, shorter papers as Research Notes. Send four identical copies of the typed material together with photocopies of illustra- tions to the Associate Editor. Do not send original handmade illustrations until the manuscript has been ac- cepted. Mail to: Petra Sierwald, Associate Editor; Division of In- sects, Dept, of Zoology, Field Museum, Roosevelt Road at Lakeshore Drive, Chicago, IL 60605 USA [Telephone: (312) 922-9410, ext. 841; FAX: (312) 663-5397; E-mail: SIERWALD@FMNH.ORG] Correspondence relating to the initial submission of a manuscript, as well as the review process, should be directed to the Associate Editor. Correspondence re- lating all other matters should be directed to the Editor. After the manuscript has been accepted, the author will be asked to submit the manuscript on a computer disc. It must be in a widely used program (preferably in MS DOS WordPerfect). Indicate clearly on the computer disc the word processing program used. FEATURE ARTICLES Title page. — The title page will include the com- plete name, address, and telephone number of the au- thor with whom proofs and correspondence should be exchanged, a FAX number and electronic mail address if available, the title in capital letters, and each author’s name and address, and the running head (see below). Abstract. — The heading in capital letters should be placed at the beginning of the first paragraph set off by a period. In articles written in English, a second abstract in an acceptable language may be included pertinent to the nationality of the author(s) or geo- graphic region(s) emphasized in the article. A second abstract in English must be included in articles not written in the latter language. Text. — Double-space text, tables, legends, etc. throughout. Three levels of heads are used. The first level (METHODS, RESULTS, etc.) is typed in capitals and on a separate line. The second level head begins a paragraph with an indent and is separated from the text by a period and a dash. The third level may or may not begin a paragraph but is italicized and sepa- rated from the text by a colon. Use only the metric system unless quoting text or referencing collection data. All decimal fractions are indicated by the period regardless of language of the text. Citation of references in the text: Cite only papers already published or in press. Include within parenthe- ses the surname of the author followed by the date of publication. A comma separates multiple citations by the same author(s) and a semicolon separates citations by different authors, e.g., (Smith 1970), (Jones 1988; Smith 1993), (Smith 1986, 1987; Smith & Jones 1989; Jones et al. 1990). Literature cited section. — Use the following style: Lombardi, S. J. & D. L. Kaplan. 1990. The amino acid composition of major ampullate gland silk (dragline) of Nephila clavipes (Araneae, Tetragnath- idae). J. Arachnol., 18:297-306. Krafft, B. 1982. The significance and complexity of communication in spiders. Pp. 15-66, In Spider Communications: Mechanisms and Ecological Sig- nificance. (P. N. Witt & J. S. Rovner, eds.). Princeton University Press, Princeton, New Jersey. Footnotes. — Footnotes are permitted only on the first printed page to indicate current address or other information concerning the author. All footnotes are placed together on a separate manuscript page. Running head. — The author surname(s) and an ab- breviated title should be typed all in capital letters and must not exceed 60 characters and spaces. The running head should be placed near the top of the title page. Taxonomic articles. — Consult a recent taxonomic article in the Journal of Arachnology for style or con- tact the Editor. Tables. — Each table, with the legend above, should be placed on a separate manuscript page. Only hori- zontal lines (usually three) should be included. Use no footnotes; instead, include all information in the leg- end. Make notations in the text margins to indicate the preferred location of tables in the printed text. Illustrations. — Address all questions concerning il- lustrations to: James W. Berry, Editor; Dept, of Biological Sci- ences, Butler University, Indianapolis, Indiana 46208 USA [Telephone (317) 940-9344; FAX (317) 940-9519; E-mail: BERRY@BUTLER.EDU] All art work must be camera-ready for reproduction. In line drawings, pay particular attention to width of lines and size of lettering when reductions are to be made by the printer. Multiple photos assembled on a single plate should be mounted with only a minimum of space separating them. In the case of multiple illus- trations mounted together, each illustration must be numbered sequentially rather than given an alphabetic sequence. Written on the back should be the name(s) of author(s) and an indication of top edge. The author should indicate whether the illustrations should be one column or two columns in width. The overall dimen- sions of original artwork should be no more than 1 1 inches (28 cm) X 14 inches (36 cm). Photocopies in review manuscripts should be reduced to the exact measurements that the author wants to appear in the final publication. Larger drawings present greater dif- ficulty in shipping and greater risks of damage for which the JOA assumes no responsibility. Make no- tations in the text margins to indicate the preferred position of illustrations in the printed text. Legends for illustrations should be placed together on the same page(s) and separate from the illustrations. Each plate must have only one legend, as indicated below: Figures 1-4. — A-us x-us, male from Timbuktu. 1, Left leg; 2, Right chelicera; 3, Dorsal aspect of geni- talia; 4, Ventral aspect of abdomen. Figures 27-34. — Right chelicerae of species of A-us from Timbuktu. 27, 29, 31, 33, Dorsal views; 28, 30, 32, 34, Prolateral views of moveable finger; 27, 28, A- us x-us, holotype male; 33, 34, A-us y-us, male. Scale = 1.0 mm. Assemble manuscript for mailing. — Assemble the separate sections or pages in the following sequence: title page, abstract, text, figure legends, footnotes, ta- bles with legends, figures. Page charges, proofs and reprints. — There are no page charges, but authors will be charged for changes made in the proof pages. Authors will receive a reprint order form along with their page proofs. Reprints will be billed at the printer’s current schedule of costs. RESEARCH NOTES Instructions above pertaining to feature articles ap- ply also to research notes, except that abstracts and most headings are not used and the author’s name and address follow the Literature Cited section. CONTENTS The Journal of Arachnology Volume 25 Feature Articles Number 1 Distribution, Movement, and Activity Patterns of an Intertidal Wolf Spider Pardosa lapidicina Population (Araneae, Lycosidae) by Douglass H. Morse 1 Notes on the Reproductive Biology and Social Behavior of Two Sympatric Species of Philoponella (Araneae, Uloboridae) by Deborah R. Smith 11 Behavior and Niche Selection by Mailbox Spiders by Robert L. Edwards and Eric H. Edwards 20 Spiders and Their Prey in Massachusetts Cranberry Bogs by Carolyn J. Bardwell and Anne L. Averill 31 Estructura Ocular de Selenops cocheleti (Araneae, Selenopidae) by Jose Antonio Corronca and Hector R. Teran 42 A New Species of Crypsidromus from Belize (Araneae, Mygalomorphae, Theraphosidae) by Steven B. Reichling 49 The Spider Family Cyatholipidae in Madagascar (Araneae, Araneoidea) by Charles E. Griswold 53 A New Species of Schizocosa from the Southeastern USA (Araneae, Lycosidae) by Gail E. Stratton 84 Research Notes The Effect of Habitat Structure on Web Height Preference in Three Sympatric Web-building Spiders (Araneae, Linyphiidae) by Marie Elisabeth Herberstein 93 On Some Camillina from Southern Africa (Araneae, Gnaphosidae) by Norman I. Platnick 97 A Useful Procedure for Estimating the Species Richness of Spiders by Robert L. Edwards 99 A Note on Euscorpius carpathicus (Scorpiones, Chactidae) from the Crimea by Victor Fet 106 The Journal of ARACHNOLOGY VOLUME 25 1997 NUMBER 2 THE JOURNAL OF ARACHNOLOGY EDITOR: James W. Berry, Butler University ASSOCIATE EDITOR: Petra Sierwald, Field Museum EDITORIAL BOARD: A. Cady, Miami (Ohio) Univ. at Middletown; J. E. Carrel, Univ. Missouri; J. A. Coddington, National Mus. Natural Hist.; J. C. Cokendolpher, Lubbock, Texas; F. A. Coyle, Western Carolina Univ.; C. D. Dondale, Agriculture Canada; W. G. Eberhard, Univ. Costa Rica; M. E. Galia- no, Mus. Argentine de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia, Missouri; C. Griswold, Calif. Acad. Sci.; N. V. Horner, Midwestern State Univ.; D. T. Jennings, Garland, Maine; V. F. Lee, California Acad. Sci.; H. W. Levi, Harvard Univ.; E. A. Maury, Mus. Argentine de Ciencias Naturales; N. I. Plat- nick, American Mus. Natural Hist.; G. A. Polis, Vanderbilt Univ.; S. E. Riechert, Univ. Tennessee; A. L. Rypstra, Miami Univ., Ohio; M. H. Robinson, U.S. National Zool. Park; W. A. Shear, Hampden-Sydney Coll.; G. W. Uetz, Univ. Cincinnati; C. E. Valerio, Univ. Costa Rica. The Journal of Arachnology (ISSN 0160-8202), a publication devoted to the study of Arachnida, is published three times each year by The American Arach- nological Society. Memberships (yearly): Membership is open to all those in- terested in Arachnida. Subscriptions to The Journal of Arachnology and American Arachnology (the newsletter), and annual meeting notices, are included with mem- bership in the Society. Regular, $30; Students, $20; Institutional, $80 (USA) or $90 (all other countries). Inquiries should be directed to the Membership Secretary (see below). Back Issues: Patricia Miller, PO. Box 5354, Northwest Mississippi Community College, Senatobia, Mississippi 38668 USA. Telephone: (601) 562- 3382. Undelivered Issues: Allen Press, Inc., 1041 New Hampshire Street, P.O. Box 368, Lawrence, Kansas 66044 USA. THE AMERICAN ARACHNOLOGICAL SOCIETY PRESIDENT: Ann L. Rypstra (1997-1999), Dept, of Zoology, Miami Univer- sity, Hamilton, Ohio 45011 USA. PRESIDENT-ELECT: Frederick A. Coyle (1997-1999), Department of Biology, Western Carolina University, Cullowhee, North Carolina 28723 USA MEMBERSHIP SECRETARY: Norman I. Platnick (appointed), American Museum of Natural History, Central Park West at 79th St., New York, New York 10024 USA. TREASURER: Gail E. Stratton, Department of Biology, University of Missis- sippi, University, Mississippi 38677 USA. BUSINESS MANAGER: Robert Suter, Dept, of Biology, Vassar College, Pough- keepsie, New York 12601 USA. SECRETARY: Alan Cady, Dept, of Zoology, Miami Univ., Middletown, Ohio 45042 USA. ARCHIVIST: Vincent D. Roth, Box 136, Portal, Arizona 85632 USA. DIRECTORS: H. Don Cameron (1997-1999), Matthew Greenstone (1997- 1999), Robert Suter (1995-1997). HONORARY MEMBERS: C. D. Dondale, W. J. Gertsch, H. W. Levi, A. F. Millidge, W. Whitcomb. Cover illustration: Photograph of Phidippus audax (Hentz) preying on a last instar noctuid corn earworm, Helicoverpa zea (Boddie). Photo by Clyde E. Morgan, USDA Agricultural Research Service (submitted by Matt Greenstone). Publication date: 21 October 1997 @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). 1997. The Journal of Arachnology 25:109-136 SALTICIDAE OF THE PACIFIC ISLANDS. II, DISTRIBUTION OF NINE GENERA, WITH DESCRIPTIONS OF ELEVEN NEW SPECIES James W, Berry: Department of Biological Sciences, Butler University, Indianapolis, Indiana 46208 USA Joseph A. Beatty: Department of Zoology, Southern Illinois University, Carbondale, Illinois 6290U6501 USA Jerzy Proszyhski: Muzeum i Instytut Zoologii PAN, ul. Wilcza 64 00-679 Warszawa, POLAND ABSTRACT. Pacific salticids of the genera Ascyltus, Bavia, Cosmophasis, Flacillula, Frigga, Ligurra, Plexippus, Thorelliola and Trite are discussed. Eleven new species are described: Ascyltus similis, Ascyltus rhizophora, Bavia fedor, Bavia sonsorol, Cosmophasis arborea, Cosmophasis lami, Cosmophasis mur- alis, Flacillula nitens, Ligurra opelli^ Thorelliola dumicola and Trite ponapensis. Illustrations and dis- tribution records are presented for the new species. Drawings of four additional species of Trite are included. This is the second in a series of papers on jumping spiders of the Pacific Islands (see Berry, Beatty & Proszyfiski 1996). In this pa- per we treat 24 species in the genera Ascyltus, Bavia, Cosmophasis, Flacillula, Frigga, Li- gurra, Plexippus, Thorelliola and Trite. Elev- en new species are described from Fiji, Samoa and the Caroline Islands. Several of the species included here have extensive distributions and have been reported many times before from the region: Ascyltus pterygodes (L. Koch 1865), Bavia aericeps Simon 1877, Bavia sexpunctata (Doleschall 1859), Frigga crocuta (Taczanowski 1878) (mostly under the name Sandalodes calvus Simon 1902), Plexippus paykullii (Audouin 1825) and Thorelliola ensifera (Thorell 1877). At least three of these familiar species, some- times more, can be found on most of the is- lands in the Pacific. Except for Wanless's (1978) revision of the genus Sobasina, very little specifically on Pa- cific salticids has been published. Zabka (1987-1995) published a series of papers un- der the general title “Salticidae of Oriental, Australian and Pacific Regions”. The empha- sis of these publications is strongly on the fau- nas of mainland Australia, Asia and the large continental islands, and little that is applicable to Micronesia and Polynesia is included. Ber- land (1934a) listed 40 salticid species from Polynesia, and in later papers (1934b, 1938, 1942) which included other Pacific areas, he added 15 more. Marples (1955a, 1955b, 1957, 1964) described six new species from Fiji, Tonga, Samoa and the Cook Islands. The New Guinea fauna described by Chrysanthus (1968) overlaps the fauna of the smaller oce- anic islands only in the case of cosmotropical or widespread Pacific species (e.g., Bavia aer- iceps Simon 1877, Menemerus bivittatus (Du- four 1831), and Plexippus paykullii (Aud. 1825)). The collections on which this paper is based were mostly made by J.W. Berry, E.R. Berry, and J.A. Beatty (indicated as JWB, ERB, and JAB in the text) in a series of collecting trips: Marshall Islands (1968, three months; 1969, three months); Palau (1973, six months); Guam, Yap, Truk, Ponape, Taiwan (1973, 1- 2 weeks each); Yap (1980, six months); Mar- quesas, Tuamotu, Society, Cook and Fiji Is- lands (1987, six months total); and Hawaii (1995, one month). Specimens borrowed from 109 110 THE JOURNAL OF ARACHNOLOGY Map 1. — Major island groups in the Pacific Ocean. the Bishop Museum (BPBM) and the Ameri- can Museum of Natural History (AMNH) were also examined and are occasionally re- ferred to in the text. As in our previous paper, we have placed new species in previously described genera to which they are most similar, recognizing that they do not always match perfectly. For ex- ample, the species of Trite illustrated (Figs. 91-104) are genitalically heterogeneous, though similar in habitus. A revision of the genus may very well dismember it and place the new species elsewhere. Species limits within genera are, again, not naiTowly defined. Small differences between Samoan and Fijian populations of Ascyltus similis new species are conceived as intraspe- cific rather than interspecific variation (Figs. 12-16, 21-28). We feel that, before additional species are described, this variation should be investigated in more specimens than we have available. None of the genera included here has been reported solely from the Pacific islands. As- cyltus, Frigga and Trite are known from the Pacific islands and from Australia {Frigga from South America, also). Flacillula, Ligur- ra and Thorelliola occur in the Pacific and in Asia (including Sri Lanka); while Bavia and Cosmophasis are found in the Pacific, Asia and Australia. Plexippus is cosmotropical. The generic diagnoses are intended to dis- tinguish only among salticid genera reported from the Pacific Islands (Micronesia and Pol- ynesia), excluding the large islands near Asia and Australia, the sub-Antarctic and the east- ern Pacific Islands. In the descriptions the genera are categorized by size as follows: small, 2-4 mm total length; medium, >4-8 mm; large, >8-16 mm; and very large, over 1 6 mm. The anterior, middle and posterior eye rows are refeiTed to, respectively, as eyes I, eyes II and eyes III. Illustrations of male palpi are of the left palp unless otherwise stated. The holotypes and other specimens of all new species will be deposited in the Bishop Museum (BPBM) (State Museum of Hawaii) in Honolulu. All adult specimens are para- types unless specifically excluded in the text; juveniles are not paratypes. Genus Ascyltus Karsch 1878 Discussion. — Seven species are currently listed in this genus (Zabka 1988; Proszynski 1990). All of these were described in the 19th or early 20th century, and most are poorly known and unrevised. Specimens mentioned in publications have almost all been identified as A. pterygodes. Species limits are not clear and the actual number of distinct species is unknown. Palpal structures are rather similar in all species. BERRY ET AL.— PACIFIC SALTICIDS 111 Diagnosis. — Distinguishable from numer- ous other fissidentate Pacific genera especially by the antero-lateral “cheek” areas of the car- apace, which are covered by iridescent scales. These areas are often broadened as well. They are usually detectable even in half-grown ju- veniles. Other diagnostic characters include the absence of lateral spines on first metatarsi, first coxae separated by more than the diam- eter of one of them, eyes in normal three rows instead of four (the second row midway be- tween the first and third), pedicel concealed by abdomen in dorsal view, and cheliceral promargin in males with a large multicusped tooth. Descriptive notes. — Medium to very large- sized fissidentate (bicuspid) salticids. Retro- lateral surface of male chelicera with stridu- latory grooves. Prolateral margin in males with a large multicusp tooth (4-6 cusps), in females a row of separate teeth. Antero-lateral portion of carapace expanded into broad, seta- fringed cheeks in males and some females. Male palps long and slender, especially the tibia; cymbium small, little wider than other palpal segments. Legs long and slender. Ceph- alothorax broad, flattened, squarish in appear- ance, usually with prominent cheeks laterally, fringed by setae and covered by reflective scales. Cheeks more pronounced in larger specimens, less so in smaller ones. Abdomen elongate, narrowing posteriorly, dorsal surface in males with a scutum, usually with indistinct edges; no scutum in females. Dorsum with two broad brown bands separated by a lighter median area; covered with adpressed colorless or brown scales. Face very low, reduced al- most to the diameter of AME, but broad, clyp- eus very low. Chelicerae in males (and in fe- male A. pterygodes) very large, diverging, elongate and broad, extended diagonally for- ward, covered with short setae. There is a bi- cuspid retrolateral tooth at midlength; a ridge on the retrolateral surface of the paturon ends in a tooth near the base of the fang. Prolateral tooth in some species preceded by an addi- tional conical tooth; fang long. In females chelicerae usually of normal size, vertical, slightly bulging basally. Legs long and brown; tibiae I and II normally with 3-3 ventral spines, plus 2 to 3 prolateral and one retrola- teral. Pedipalps thin and long. Ascyltus pterygodes (L. Koch 1865) Figs. 1, 4, 7, 9, 10, 11; Map 2 Hyllus pterygodes L. Koch 1865 Ascyltus pterygodes (L. Koch): Karsch 1878 Diagnosis. — Large to very large speci- mens; chelicerae large and divergent in both sexes. Male: Tibia of palp clearly longer than cymbium, fang furrow with transverse ridges, no diagonal ridge on anterior surface of che- licera. Female: Internal ducts of epigynum long, but not extending forward beyond “win- dows” (Figs. 10, 11), cheeks well developed. Description. — Male: {n = 5). Total length 13-19 (x = 15.4), length of carapace 5. 5-7.0 (x = 6.20), maximum carapace width 5. 1-6.9 (x = 5.94), eye field length 2.3-3. 1 (x = 2.77), eye row I width 3. 1-4.0 (x = 3.58). Cephalothorax brown, with eye field and ven- tral margins dark brown; cheeks very large and prominent. Abdomen brown with darker marginal streaks, sides brownish-grey, spin- nerets dark greyish-brown. Frontal aspect- eyes I surrounded dorsally by orange setae, ventrally by white, chelicerae particularly broad, blackish brown, legs brown anteriorly. One bicusp retrolateral cheliceral tooth plus a conical distal one offset somewhat from the fang furrow, one large five-cusped prolateral cheliceral tooth. Ventral aspect: mouth parts brown, endites with antero-external edges ex- panded triangularly, sternum light brown, me- dially lighter. Legs: Leg formula 1 -2-4-3, pa- tella-tibia III equal to IV. Patella-tibia I length 6.4-9. 5 (x == 7.84). Coxae brown anteriorly and yellow posteriorly; abdomen ventrally greyish with brownish scales. Palp: With tibia distinctly longer than the cymbium (see Fig. 1). Female: (n 5). Total length 16-20 mm (x = 17.1), length of carapace 5. 5-7.0 (x = 5.98), maximum carapace width 5. 1-6.5 (x = 5.64), eye field length 2. 7-3. 2 (x = 2.81), eye row I width 3. 5-4.0 (x = 3.69). Cepha- lothorax brown like male, with eye field and ventral margins dark brown; cheeks very large and prominent, without vertical horn-like tufts of stiff black bristles near eyes II; chelicerae protruding forward, somewhat diverging, long (about Vi length of cephalothorax) and broad. Abdomen oval, swollen medially and narrow- ing posteriorly, but without scutum, covered densely with brownish scales on white back- THE JOURNAL OF ARACHNOLOGY Figures 1-10. — Comparison of species of Ascyltus. 1-3. Ventral view of left palps all drawn to the same scale. 1, A. pterygodes; 2, Ascyltus similis new species; 3, Ascyltus divinus. 4-6. Abdominal patterns of females of Ascyltus, (drawn to different scales); 4, A. pterygodes', 5, A. similis', 6, A. rhizophora', 7, Dorsal view of cephalothorax of female Ascyltus pterygodes', 8, Frontal view of A. rhizophora new species; 9, Ascyltus pterygodes (L. Koch) epigynum; 10, Internal structure of A. pterygodes epigynum — single spermatheca and ducts. ground, with two broad dark brown streaks of scales along posterior % of abdomen, divided by a naiTow gap into two blocks; sides with mosaic of fine, dense brownish and whitish streaks of scales. Spinnerets brownish. Frontal aspect: much like male, eyes I suiTounded dorsally and ventrally by orange setae, chelic- erae blackish-brown covered with short sparse whitish setae. One bicusp retrolateral chelic- eral tooth, six prolateral cheliceral teeth. Ven- tral aspect: as in male, except endites with an- tero-external edges rounded, sternum light brown, medially lighter; coxae brownish yel- low; abdomen ventrally very light brownish. Legs: Leg formula 4-3- 1-2, patella-tibia III equal to IV. Patella-tibia I length 5. 1-7.8 (x = 6.14). Legs light brown, the first pair darker. Epigynum: With septum nan'ow at mid-length, internal duct with double loop about half as long as diameter of “window” or slightly more (see Figs. 9, 10). Material examined. — FIJI: Viti Levu, Nan- darivatu, in house. Id, 11 April 1987 (JAB); Nan- darivatu, along stream near swimming pool, Id, 12 April 1987 (JWB); Nandarivatu, on shrub at swim- ming pool. Id, 12 April 1987 (JAB); Nandarivatu, at swimming pool, 1 $, 14 April 1987 (JAB); Nan- darivatu, in house. Id, 17 April 1987 (JAB). HA- WAII: Hawaii County, Captain Cook, on bush in nursery. Id, 15 January 1988; Manuka State Park, mesic forest, elev. 1770 ft., 2 9 13imm, 11 February 1995; Kalopa State Park, shaking banana leaves, elev. 2500 ft., Id 3 98imm; Waipio Valley Lookout, on clay bank. Id, 14 February 1995; Lapahoehoe, elev. 500 ft., shaking banana leaves, Id 12imm, 20 Febmary 1995; Lapahoehoe, elev., 1100 ft., along gulch. Id llimm, 20 February 1995; Isaac Hale Beach Park, Pandanus litter, 1 9 2imm, 23 February 1995. (All Hawaiian specimens collected by J.W. & E.R. Ben-y.) Distribution. — Reported from Hawaii, Sa- moa, Fiji, Tonga, Niue, the Ellice, Tokelau, New Hebrides, Loyalty and Society Islands. Ascyltus divinus Karsch 1878 Figs. 3, 14, 17-20; Map 2 Ascyltus simplex Karsch 1878: synonymized by Zabka (1988) Discussion. — Zabka’s (1988) drawing of the male chelicerae shows a simple prolateral tooth, not the multicusp tooth typical of the BERRY ET AL.— PACIFIC SALTICIDS 113 Figures 11-16. — Comparison of chelicerae in males of several Ascyltus species. 11, Ventral view in Ascyltus pterygodes from Fiji; 12, Dorsal view in Ascyltus similis new species from Samoa; 13, Ventral view in Ascyltus similis new species from Samoa; 14, Ascyltus divinus from Fiji [cf. Zabka 1988: 424-427, fig. 8]; 15, Dorsal view of chelicera of Ascyltus similis new species from Fiji; 16, Ven- tral view of chelicera in A. similis new species from Fiji. genus. However, our specimens show a row of small cusps on one edge of the large tooth. These could easily be overlooked from some angles of view and may have been further ob- scured by setae in the dry specimens Zabka examined. Other characters of our male spec- Map 2. — Distribution of four species of Ascyltus in the Pacific. Ascyltus pterygodes (★), Ascyltus di- vinus (A), Ascyltus similis new species (□), Ascyl- tus rhizophora new species( 0 ). imens match Zabka’s description and illustra- tions well. Apparent differences in the females are attributed to individual variation, different styles of drawing and differences in the con- dition of the specimens.. As the males of no other species have an anterior ridge on the chelicera, we include our specimens in A. di- vinus. Diagnosis. — Male chelicerae with sparse inconspicuous setae dorsally, with prominent diagonal sclerotized ridge, cusps of prolateral tooth small, running down one edge of tooth. Fang furrow with transverse ridges. Internal ducts of epigynum short, cheeks of female not expanded or rimmed by long setae. Description. — Male: {n = 3). Total length 7. 9-8. 6 (x = 8.19), length of carapace 3.3- 3.6 (x = 3.50), maximum carapace width 2.9- 3.4 (x = 3.08), eye field length 1. 7-2.0 (x = 1.91), eye row I width 2.4-2. 7 (x = 2.54). Cephalothorax yellowish-brown with eye field light brown, cheeks present but not prominent. Abdomen brownish-yellow, sides whitish. Figures 17-20. — Ascyltus divinus Karsch. 17, Epigynum; 18, Internal structure of epigynum showing single spermatheca and ducts. 19, Left palp, ventral view; 20, Left palp, lateral view. 114 THE JOURNAL OF ARACHNOLOGY spinnerets light brown. Frontal aspect-eyes I suiTounded by indistinct whitish setae, some with slightly orange shade, chelicerae with prominent black sclerotized ridge on anterior surface, anterior legs yellow. One proximal bi- cusp, plus two low conical (one proximal to bicusp and one distal to bicusp) retrolateral teeth; one large 5-cusped prolateral cheliceral tooth (plus another conical one more distal). Ventral aspect: mouth parts brown, endites with antero-external edges rounded, sternum light brown, medially lighter; coxae yellow; abdomen ventrally whitish with four indistinct longitudinal lines of brownish spots. Legs: Leg formula 1-2-4= 3, patella-tibia III equal to IV. Patella-tibia I length 4. 3-6.0 (x = 5.23). Legs yellow, spination of tibia I differs from the remaining species by lateral spines being only indistinctly shortened and presence of similar spines on retro-lateral surface as well, the same spination appears on tibia II. Palpal structures not distinctive. Female: {n = 5). Total length 7. 5-8. 3 (x = 7.91), length of carapace 2. 6-3.4 (x = 3.21), maximum carapace width 2.0-2. 8 (x = 2.55), eye field length 1.3-1. 8 (x = 1.67), eye row I width 2.0-2.4 (x = 2.27). Differs from other Ascyltus by absence of distinct cheeks, the cephalothorax here is as broad as under eyes III, but not broader. Cephalothorax with eye field fawn, an indistinct pattern of longer and broader whitish scales along midline, along posterior edge and along lateral eyes, sur- rounding two spots of light brownish scales; rims of eyes I covered dorsally with longer whitish scales, a few longer orange scales be- tween AME. A whitish diamond-shaped area behind eye field followed by slightly darker, light fawn middle thorax with sparse orange setae, lower thorax and sides whitish. Abdo- men whitish, with sparse widely spaced or- ange scales and also scattered dai*k, upright short bristles; no pattern visible. Frontal as- pect: eyes I suiTounded with white setae with an indistinct dot of orange setae laterally and medially, AME suiTounded by light brown area, very thin laterally and ventrally, whitish area under ALE, expanded laterally but not making any extended plate, rimmed dorsally by a diagonal line of small whitish scales, fol- lowed laterally by a line of red scales running from ALE sidewards. Clypeus with thin line of horizontal whitish setae. Chelicerae vertical and slightly diverging, bulging basally, yel- low, their length being about twice diameter of AME. One bicusp retrolateral cheliceral tooth, four prolateral cheliceral teeth. Pedi- palps whitish, with long white setae. Ventral aspect: generally whitish, chelicerae and mouth parts yellow. Legs: Leg formula 4-3- 1-2; patella-tibia III equal to IV. Patella-tibia 1 length 2. 0-2. 9 (x = 2.60). Legs yellowish dorsally with prominent brown spines, ven- trally whitish. Epigynum: With septum broad at mid-length, internal duct short, inconspic- uously looped (see Figs. 17, 18). Material examined. — FIJI: Viti Levu, mangrove swamp by road near Namuka Harbor, sweeping, 19.2 May 1987 (JWB & ERB). Nandarivatu, elev. 900 m, tree shaking in shrubs, 19, 11 April 1987 (JWB & ERB). Namosi Road, 7.7 km north of Queen’s Road, roadside sweeping & shaking, 2 92imm, 7 May 1987 (JAB, JWB & ERB). Lo- maivuna Distr., 3 km N of Nanggali, tree shaking, in pine, 1319 limm, 30 May 1987 (JWB & ERB). Namosi District, hilltop forest about 7 km N of Queen’s Road on Namosi Road, 13, 19 May 1987 (JWB & ERB). Nausori Highlands, forest reserve Koronsingalevu Block, elev. 1500 ft., sweeping & shaking, 13, 27 May 1987 (JWB & ERB). Distribution. — Reported only from Fiji and Australia (Zabka 1988). Ascyltus similis new species Figs. 2, 5, 12, 13, 15, 16, 21-28; Map 2 Holotype. — Male from Fiji: Viti Levu, 7 mi. N of Singatoka, sweeping and shaking bushes along river bank, 21 May 1987. (J.W. & E.R. Ben-y) (BPBM). Etymology. — The name similis, similar, re- fers to the resemblance of this species to A. pterygodes. Diagnosis. — Smaller than A. pterygodes, which it resembles. Male chelicerae lacking dorsal sclerotized ridge, fang fuiTOw with only slight transverse ridges. Females with seta- rimmed cheeks. Epigynum with short internal ducts. Tibia of male palp only a little longer than cymbium. Description. — Male: (n = 3). Total length 8.0-11.5 (x = 9.57), length of carapace 4.4- 5.2 (x = 4.71), maximum carapace width 3.9- 4.5 (x = 4.08), eye field length 2. 3-2. 5 (x = 2.38), eye row I width 2. 9-3.0 (x = 2.96). Cephalothorax brown, with eye field and ven- tral margins dark brown, cheeks very large and prominent. Abdomen with scutum brown, margins and sides lighter brownish grey. BERRY ET AL.— PACIFIC SALTICIDS 115 Figures 21-28. — Ascyltus similis new species. Comparison of left palps and epigyna for specimens from Fiji and Samoa. Comparable structures are drawn to the same scale. 21, Palp, ventral view (Fiji); 22, Palp, ventral view (Samoa); 23, Palpal tibia , lateral view (Fiji); 24, Palpal tibia, lateral view (Samoa); 25, Epigynum (Fiji); 26, Internal structure of epigynum (Fiji); 27, Epigynum (Samoa); 28, Internal struc- ture of epigynum (Samoa). Frontal aspect: eyes I surrounded dorsally by orange setae, ventrally by white; chelicerae broad, blackish-brown, legs brown anteriorly. Retrolateral cheliceral teeth; one bicusp, plus two low rounded bumps, one proximal and one medial to the bicusp tooth; one four-cus- ped prolateral cheliceral tooth. Ventral aspect: mouth parts brown, endites with antero-exter- nal edges expanded triangularly, sternum light brown, medially lighter; coxae brownish-yel- low; abdomen greyish ventrally with brownish scales. Legs: Leg formula 1 -2-4-3; patella-tib- ia III equal to IV. Patella-tibia I length 4.9- 5.7 (x = 5.27). Palp: Structures as in A. pter- ygodes except for proportionately shorter tib- ia. Female: {n = 5). Total length 9.8-12.6 (x = 11.40), length of carapace 4. 1-5.3 (x = 4.79), maximum carapace width 3. 3-4. 5 (x = 3.96), eye field length 2.0-2. 7 (x = 2.44), eye row I width 2.7-3.4 (x = 3.08). Cephalotho- rax as brown as in male, with eye field and ventral margins dark brown; cheeks large and prominent, vertical hom-like tufts of stiff black bristles near eyes II, chelicerae not pro- truding forward. Abdomen narrowing poste- riorly, without scutum, with remnants of two intensely brown lateral streaks of narrow scales along posterior % of abdomen, extend- ed anteriorly by greyish scales, divided into three blocks by narrow gaps, median longi- tudinal area whitish with colorless scales; sides whitish covered with colorless scales. Spinnerets brownish. Frontal aspect: eyes I surrounded dorsally by orange setae, no sur- rounding setae ventrally, but with a few dark setae along edge of clypeus, and dense short whitish setae on bases of chelicera make a white line under AME; chelicerae bulging ba- sally but directed vertically, much smaller than in male, light brown, covered with short and sparse whitish setae. One bicusp retrola- teral cheliceral tooth, four prolateral cheliceral teeth. Ventral aspect: mouth parts brown, en- dites with antero-external edges rounded, ster- num yellowish; coxae yellowish; abdomen ventrally whitish. Legs: Leg formula 1-3 =4-2; patella-tibia III shorter than IV. Patella-tibia I length 3. 6-5. 2 (x = 4.51). Legs brown ante- 116 THE JOURNAL OF ARACHNOLOGY Figures 29-30. — Ascyltus rhizophora new spe- cies from Fiji. 29, Epigynum; 30, Internal structure of epigynum showing single spermatheca and ducts. riorly. Epigynum: See diagnosis and Figs. 25- 28. Material examined. — FIJI: Viti Levu, Suva, Lami Beach, on shrub foliage, 1$, 3 May 1987 (JAB & ERB). Seven mi. N of Singatoka, sweeping & shaking bushes along river bank, 1 (3 1 9 , 21 May 1987 (JWB & ERB); AMERICAN SAMOA: Tu- tuila, Fagatogo, 23 1 9 limm, 13 July 1973 (JAB); 3 95imm, 14 July 1973 (JAB). Distribution. — Known only from Fiji and American Samoa. Ascyltus rhizophora new species Figs. 6, 29; Map 2 Holotype. — Female from Fiji: Viti Levu, near Namuka Harbor, mangrove swamp, sweeping, 2 May 1987, (J.W. & E.R. BeiTy) (BPBM). Etymology.- — A noun in apposition after the mangrove genus Rhizophora. Diagnosis. — Female with long internal epi- gynal ducts that loop forward beyond anterior margin of “windows”. With seta-rimmed cheeks. Description. — Male: Male is unknown. Female: {n — 1). Total length 8.7, length of carapace 4.0, maximum carapace width 3.2, eye field length 1 .9, eye row I width 2.6. Re- sembles other Ascyltus by small cheeks, ex- tended by rims of bent setae; rims of eyes I covered dorsally with reddish setae. Cepha- lothorax whitish, with eye field anteriorly dark brownish-grey. Very indistinct, small, trans- parent adpressed scales, colorless and light brown; posterior median whitish area on eye field is continued as median broad whitish belt along the whole length of thorax, limited on both sides by broad darker belts, consisting of small grey spots and covered with semi-trans- parent brownish scales; sides whitish covered in upper parts by sparse brownish scales, low- er sides whitish, limited by the thin dark brown line on the ventral edge. Abdomen with distinct reddish-brown pattern on whitish background (Fig. 6). Frontal aspect: eyes I surrounded with orange-red setae, face ap- pears light greyish-brown, due to scales and setae on pale tegument; cheek plates small but broader than cephalothorax behind them, cov- ered by adpressed brownish-grey, shiny setae, rimmed laterally by a thin indistinct line of white setae, dorsally by a diagonal line of red scales mnning from ALE sidewards. Clypeus obsolete, with a thin line of horizontal whitish setae. Chelicerae vertical and slightly diverg- ing, bulging basally, yellow, their length being about twice diameter of AME. Pedipalps whit- ish, with long white setae and a spot of brownish scales on patella. One retrolateral cheliceral tooth, four prolateral cheliceral teeth. Ventral aspect: generally whitish, che- licerae and mouth parts yellow. Legs: Leg for- mula 4-3- 1-2, patella-tibia III equal to IV. Pa- tella-tibia I length 3.8. Legs dorsally yellowish with indistinct darker rings and prominent brown spines; ventrally whitish. Epigynum: See diagnosis and Figs. 29, 30. Material examined. — FIJI; only the holotype. Distribution. — Known only from Fiji. Genus Bavia Simon 1877 Discussion. — Proszynski (1990) catalogs 12 species of Bavia, occurring from the Philippines and southeast Asia to Australia. Five of these have recently been discussed by Zabka (1988), who questions the placement of B. annamita Simon 1903 and B. thorelli Simon 1903 in the genus. We describe two additional species. Diagnosis. — Distinguishable from the few other pluridentate genera in the Pacific {Lag- nus L. Koch 1879, Myrmarachne MacLeay 1838 and Lepidemathis Simon 1903) by hav- ing the pedicel concealed by anterior part of abdomen, coxae II and III not more widely spaced antero-posteriorly than other coxae, cephalothorax low, relatively flat and strongly convex laterally, ocular quadrangle parallel- sided, cheliceral retromargin with 6-7 small, acute contiguous teeth, promargin with three larger teeth, the middle one the largest. Descriptive notes. — Medium-to-large plur- identate salticids with low broad carapace. BERRY ET AL.— PACIFIC SALTICIDS 117 Figures 31-34. — Bavia aericeps. 31, Left palp ventrally; 32, Palp laterally; 33, Epigynum; 34, Internal structure of epigynum showing left spermatheca and ducts. elongate, narrow abdomen broadest anteriorly and narrowing to posterior end, with long spinnerets. Endites much longer than wide, of- ten abruptly expanded distally. First pair of legs longest, more robust and darker in color than other legs. Tibia with ventral spines in two rows of three each, metatarsi with 2-2 ventral spines in distal half. First legs and car- apace reddish-brown, with spots or short streaks of light colored scales on carapace. Other characters as in diagnosis and species descriptions. Bavia aericeps Simon 1877 Figs. 31-34, 45; Map 3 Bavia aericeps Simon 1877: Zabka 1988b. Acompse suavis L. Koch 1879: Keyserling 1883. Description. — Male: (n = 5). Total length 8.4-11.3 (x = 10.16), length of carapace 3.5- 4.9 (x == 4.14), maximum carapace width 2.4- 3.9 (X - 3.10), eye field length 1. 6-2.3 (x - 1.98), eye row I width 2. 0-2. 6 (x = 2.24). Cephalothorax reddish-brown, lighter dorsal- ly, eye field dark brown, followed by trans- verse whitish spot, a few small colorless setae along lateral eyes and a row of long brown and reddish bristles above eyes I; posterior slopes of thorax and sides with few erect black setae, a few white spots on posterior thoracic slope. Abdomen with three longitu- dinal streaks: the median one whitish, the two marginal ones darker, weakly brownish-violet in alcohol, followed by three pairs of small spots; a thin marginal whitish line anteriorly, expanding into whitish sides; spinnerets light brownish-grey. Face brown with dense white clypeal “mustache”, eyes I surrounded by in- distinct reddish setae. Diameter of AME = 2.5 diameters of ALE. Six retrolateral cheliceral teeth, three prolateral cheliceral teeth. Legs: Leg formula 1 -4-2-3, patella-tibia III shorter than IV. Patella-tibia I length 3. 0-5. 2 (x = 3.96). Femur and tibia I with inconspicuous spots of whitish setae prolaterally in middle of patella and apex of tibia. Tarsus I light yel- low. Endites elongate with a rectangular elon- gate expansion along external edge. Chelic- erae posteriorly brown, endites, labium and anterior coxae light greyish-brown, coxae III-IV whitish-yellow; sternum yellow, brown rimmed; abdomen ventrally greyish-white with light brown, sclerotized epigastric fold, grey rectangular area in the posterior third of abdomen, spinnerets surrounded by a narrow dark ring. Palp: Embolus short and straight (Figs. 31, 32). Pedipalps light brownish-grey, without contrasting spots, with longer dark se- tae along prolateral edge of cymbium and tib- ia. Female: (n = 5). Total length 9.7-12.5 (x = 10.70), length of carapace 4. 3-4. 9 (x = 4.62), maximum carapace width 3. 3-3. 9 (x = 3.56), eye field length 2. 1-2.4 (x = 2.20), eye row I width 2. 3-2. 6 (x = 2.40). Colors as in male, except behind eye field there is a whit- ish transverse area with fovea in the middle, pedipalps with flattened dorsal surfaces, yel- low, medially dark brown, framed laterally with dense fringes of white, short setae. Six retrolateral cheliceral teeth, three prolateral cheliceral teeth. Endites elongate, externally broadened and rounded, without depression or expansion. Legs: Leg formula 1 -4-2-3, patel- la-tibia III shorter than IV. Patella-tibia I length 3. 3-3. 8 (x = 3.57). Epigynum: With openings widely separated, posterior margin 118 THE JOURNAL OF ARACHNOLOGY Map 3. — Distribution of four species of Bavia in the Pacific. Bavia aericeps (★), Bavia sexpunctata (■), Bavia fedor new species (□), Bavia sonsorol new species (A). projecting in midline, internal duct relatively wide (Figs. 33, 34). Material examined. — COOK ISLANDS: Rar- otonga, Muri, on taro leaf in yard, 1 $ , 24 March 1987 (JAB). Tupapa Valley, on grass. Id, 2 April 1987 (JAB). Taakoka Island, tree shaking, ld396imm, 23 March 1987 (JWB & ERB). Aror- angi. Are Metua at Rutaki Road, 1 d2imm, 9 March 1987 (JWB & ERB). Arorangi village, elev. 30 m, tree shaking, ld4imm, 14 March 1987 (JWB & ERB). Arorangi village, on house. Id, 9 March 1987 (JWB & ERB). Avarua, 0-100 m, I95imm, August 1979 (N.L.H.Kiauss) (BPBM). Avarua, 0- 100 m, 3dlimm, August 1979 (N.L.H.Krauss) (BPBM). Titikaveka, 0-100 m, I9imm, October 1977 (N.L.H.Krauss) (BPBM). Aitutaki, Tautu, tree shaking, I92imm, 26 March 1987 (JAB & JWB). AMERICAN SAMOA: Tutuila, Fagatogo, shaken from dead lower leaves of bananas, I93imm, 14 July 1973 (JAB). Fagatogo, shaken from dead low- er leaves of bananas, ld2imm, 13 July 1973 (JAB). FIJI: Viti Leva, about 5 mi. W of Nausori, Nan- duruloulou Res. Station, Id 1 92imm, 15 May 1980 (JAB). SOCIETY ISLANDS: Moorea, Paopao vil- lage, litter, elev. 100 m, I92imm, 11 January 1987 (JWB & ERB). Huahine, Fare, 1 $2imm, February 1961 (N.L.H.Krauss) (BPBM). Fare, 0-100 m. Id, March 1972 (N.L.H.Krauss) (BPBM). Raiatea, Utotoa, 0-100 m, 1 9, March 1972 (N.L.H.Kiauss). MARQUESAS ISLANDS: Hiva Oa, above Atuo- na, elev. 500 m, sweeping and shaking vegetation, ld3imm, 12 February 1987 (JWB & ERB). AUS- TRAL ISLANDS: Rurutu, Moerai, 0-150 m, ld2 9limm, October 1977 (N.L.H.Krauss) (BPBM). GILBERT ISLANDS: Pacific Sci. Bd., 1 d (E.T. Moul) (BPBM). Distribution. — Sumatra, New Hebrides, Fiji, Samoa, Marquesas, Hawaii, New Guinea, Marianas, Ellice, Austral, Gilbert, Cook and Society Islands. Bavia sexpunctata (Doleschall 1859) Figs. 35-38, 46; Map 3 Salticus sexpunctatus Doleschall 1859. Bavia sexpunctata: Thorell 1890; Proszynski 1984; Zabka 1988. Acompse dulcinervis L. Koch 1879: Thorell 1881. Description. — Male: {n = 5). Total length 8.1-10.8 (x = 9.14), length of carapace 3.4-4.2 (x = 3.74), maximum carapace width 2.4-3. 4 (x = 2.96), eye field length 1. 8-2.2 (x = 1.96), eye row I width 2.0-2.5 (x = 2.24). Cephalo- thorax reddish-brown, gradually lighter dorsally, eye field dark brown, followed by broad trans- verse lighter area with a few whitish setae, a few small whitish scales along lateral eyes and a spot of white setae above junction of AME; posterior slopes of thorax and sides with few small erect black setae, sparse smaller brown ones, and a few in'egulai' lines of white setae and scales. Abdomen with median ai'ea whitish limited by several pairs of elongate spots, lateral to which are many short narrow violet-brown spots, lower sides whitish, spinnerets light yel- lowish-brown. Face brown with white setae along vential edge of clypeus, with triangulai' median patch of whitish setae above AME, and suiTounded with inconspicuous reddish setae. ALE almost touching AME, diameter of AME = 2 diameters of ALE. Apical pait of cymbium and prolateral pait of its basal half whitish-yel- low, reti'olateral basal part brownish, tibia retro- laterally blackish, prolaterally whitish-yellow, patella and femur brown with mai'ginal rows of short white setae. Seven retrolateral cheliceral teeth, three prolateral cheliceral teeth. Endites elongate with small triangular expansion pointed anteriorly. Endites, labium, anterior coxae and sternum brown, coxae II yellowish-brown, cox- ae III-IV whitish-yellow; abdomen ventrally gieyish-white with light brown, sclerotized ep- igastiic fold, long light grey rectangulai' area in the posterior third of abdomen, no daik ring around spinnerets. Legs: Leg formula 1 -4-2-3; patella-tibia III shorter than EV. Patella-tibia I length 3. 5-4.9 (x = 4.02). Inconspicuous spots of whitish setae prolaterally in middle patella and apical tibia. Tai'sus I light yellow. Palp: With embolus broad and sickle- shaped, bulb with proximal bifurcation (Figs. 35, 36). Female: {n = 5). Total length 10.0-12.0 (x = 1 1 .04), length of carapace 4. 0-4. 5 (x = BERRY ET AL.— PACIFIC SALTICIDS 119 Figures 35-38. — Bavia sexpunctata. 35, Left palp ventrally; 36, Palp laterally; 37, Epigynum; 38, In- ternal structure of epigynum showing showing right spermatheca and ducts. 4.20), maximum carapace width 3. 2-3. 7 (x = 3.32), eye field length 2. 1-2.3 (x = 2.20), eye row I width 2. 3-2. 5 (x = 2.36). Behind eye field there is a light orange transverse area with fovea in the middle, with minute white setae; slopes of thorax and sides covered with sparse white setae. Face brown with line of longer, white setae along ventral edge of clyp- eus. Chelicerae bulging basally. Seven retro- lateral cheliceral teeth, three prolateral chelic- eral teeth. Pedipalps with flattened dorsal surfaces, yellow, medially dark brown, framed laterally with dense fringes of white, short se- tae. Endites elongate, externally rounded, without depressions or expansions. Chelicerae posteriorly, endites, labium, anterior coxae and sternum brown, coxae II yellowish- brown, coxae III-IV whitish yellow; abdomen ventrally greyish-white with light grey rect- angular area in the posterior third of abdomen. A very small and inconspicuous protuberance with dark setae in front of spinnerets, no dark ring around spinnerets. Legs: Leg formula 1 =4-2-3; patella-tibia III shorter than IV. Pa- tella-tibia I length 3.4-3. 8 (x = 3.52). Epi- gynum: With large oval depressions separated by a septum much narrower than in B. aeri- ceps, internal ducts narrow (Figs. 37, 38). Material examined. — CAROLINE ISLANDS: Palau Islands, Arakabesan L, mixed forest, shaken from trees, 50-75 ft. el., Idlimm, 16 February 1973, (JWB). Arakabesan I., in fallen betel palm fronds, l$6imm, 23 March 1973 (JWB & JAB). Malakal L, shaken from fallen palm fronds, l$4imm, 8 March 1973 (JWB & JAB). Koror I., shaken from trees at Japanese temple ruins, 1 9 limm, 17 March 1973 (JWB & JAB). Koror I., shaken from banana trees, 29 limm, 21 March 1973 (JAB & JWB). Koror I., taro patch, 2349l7imm, 7 March 1973 (JAB & JWB). Koror L, near taro patch, from nipa palm inflorescences, 29, 9 May 1973 (JAB & JWB). Babelthuap I., Airai, low trop- ical forest N of airstrip, 2(35 9l4imm, 27 March 1973 (JAB & JWB). Babelthuap I., Airai, from fall- en betel palm fronds, 2 9 8imm, 11 March 1973 (JAB & JWB). Babelthuap L, Airai, tree shaking near SDA school, 1329,41 March 1973 (JAB & JWB). Babelthuap L, Ngaremlengui, forest, 139imm, 21 April 1973 (JAB & JWB). Peleliu I., mixed tropical forest, 29l3imm, 22 March 1973 (JWB & ERB). Angaur I., shaken from trees, mixed tropical forest, I93imm, 30 April 1973 (JAB & JWB). Angaur I., banana-betel palm community, 432915imm, 27 April 1973 (JAB & JWB). Tobi I., shaken from trees, forest, I923imm, 8 April 1973 (JWB & ERB). MARSHALL ISLANDS: Kwajalein Atoll, Ennylebegan, shaken from trees, 19, 21 July 1969 (JWB). Kwajalein Atoll, Gugee- gu, shaken from trees, I92imm, 24 July 1969 (JWB). Distribution. — India to Australia, Caroline Islands, Marshall Islands. Bavia fedor new species Figs. 39-42, 47; Map 3 Holotype. — Male from Caroline Islands, Yap, Yap Island, Fedor village, in taro patch, 11 February 1980 (J.W. Berry) (BPBM). Etymology. — This species is named in honor of the people of Fedor village in the Yap Islands where it was collected. Diagnosis. — Male with a distinct rounded dorsal tibial apophysis on the palp, in addition to the lateral apophysis; female epigynum with broad septum between openings (Fig. 41). Other species lack dorsal apophysis on male palp (Figs. 45-48) and have distinctly different epigyna. Description. — Male: {n = 3). Total length 120 THE JOURNAL OF ARACHNOLOGY Figures 39-42. — Bavia fedor new species from Yap. 39, Left palp ventrally; 40, Palp laterally; 41, Epigynum; 42, Internal structure of epigynum showing left spermatheca and ducts. 6. 0-7. 5 (x = 6.75), length of carapace 2.7- 3.9 (x = 3.26), maximum carapace width 2.0- 3.0 (x = 2.50), eye field length 1.3-1. 9 (x = 1.66), eye row I width 1.6-2. 2 (x = 1.90). Cephalothorax resembles Bavia aericeps ex- cept: face light brown with sparse greyish- brown setae, no contrasting line of white setae along ventral edge of clypeus, ALE aligned Va of their diameter above dorsal rim of AME, almost touching them, diameter of AME = 2 diameters of ALE, pedipalps brownish-yel- low, with a peculiar plate-like dorsal process on palpal tibia, apart from lateral apophysis, prominent but isolated bunches of longer greyish setae prolaterally on cymbium, tibia and patella. Endites elongate, broader apically and nan-owing basally, with oval depression on the external apical edge followed by semi- circular expansion. Six retrolateral cheliceral teeth, three prolateral cheliceral teeth. Legs: Leg formula 1— 4-2-3; patella-tibia III shorter than IV. Patella-tibia I length 2.2-4. 1 (x = 3.10). Leg I light brown with tarsus I whitish, inconspicuous spot of whitish setae prolater- ally on patella; tarsus I white, remaining leg I brownish-yellow, denser and longer grey setae ventrally on patella-tibia-metatarsus I and along ventro-retrolateral edge of femur I; no such setae on legs II-IV, which are lighter, brownish-yellow. Palp: With bulb deeply notched proximally as in B. sexpunctata. Em- bolus long, slender and curved; attached near distal end of bulb. Female: {n = 2). Total length 9.5-10.6 (x = 10.05), length of carapace 3. 8-4. 6 (x == 4.20), maximum carapace width 3. 0-3. 8 (x = 3.40), eye field length 1.9-2. 2 (x = 2.05), eye row I width 2. 2-2.4 (x = 2.30). Six retrola- teral cheliceral teeth, four prolateral cheliceral teeth. Legs: Leg formula 1— 4-2-3, patella-tib- ia III shorter than IV. Patella-tibia I length 3. 1-3.9 (x = 3.50). Coloration essentially as in male. Epigynum: See diagnosis and Figs. 41, 42). Material examined.— CAROLINE ISLANDS: Yap, Fedor Village, on coconut fronds, 19, 12 March 1980 (JWB). Fedor Village, taro patch. Id, 1 1 February 1980 (JWB). Fanif, on nipa palm pet- iole, 19, 14 April 1980 (JWB). Colonia, near house. Id, 19 March 1980 (JWB). Gagil-Tomil, shaking banana leaves, 19, 29 May 1973 (JAB & JWB). Fais Island, (no date), 1 9, E12087 (BPBM). Distribution. — Known only from Yap and Fais in the Caroline Islands. Bavia sonsorol new species Figs. 43, 44, 48; Map 3 Holotype. — Male from Caroline Islands, Sonsorol Island, mixed tropical forest, 6 April 1973 (J.W. & E.R. Ben-y) (BPBM). Etymology. — This species is named for the island of Sonsorol in the Palau Islands where it was collected. Diagnosis. — Palp with broadly triangular lateral tibial apophysis which, in the only available specimen, is transparent and difficult to see. Lateral apophysis of the other species naiTower basally (Figs. 43, 44, 45-48). Description. — Male: (n = 1). Total length 8.1, length of carapace 3.5, maximum cara- pace width 3.2, eye field I length 1.6, eye row I width 1.9. Cephalothorax oval, broadening at the level of eyes III, brown, lighter dorsally, eye field fawn, followed by yellowish area be- hind, posterior slope of thorax dorsally lighter brown, yellowish at the hind margin. Cepha- BERRY ET AL.— PACIFIC SALTICIDS 121 Figures 43, 44. — Bavia sonsorol new species from Sonsorol, Palau District. 43, Left palp ven- trally (bulb collapsed); 44, Palp laterally. lothorax almost bare, with whitish setae on thoracic slope and very dense whitish setae beneath and surrounding lateral eyes; a row of sparse short, indistinct setae diagonally from ALE along the crest of broadened area, some- what reminiscent of cheeks in Ascyltus; a row of long colorless and yellowish horizontal se- tae above eyes 1. Abdomen in alcohol whitish. Spinnerets yellowish-brown. Face light red- dish-brown. Thick lines of white setae on the reddish-brown chelicerae: one vertical along prolateral edge, the second, transverse across basal part of chelicerae. ALE aligned Vs of their diameter above dorsal rim of AME, al- most touching them, diameter of AME = 0.7, diameters of ALE = 0.3 mm. Endites elon- gate, semicircularly broader apically and nar- rowed basally, without oval depression on the external apical edge or any separate expan- sion. Chelicerae, endites, and labium brown, anterior coxae light brown, coxae II-IV whit- ish-yellow; sternum pale yellow, brown rimmed; abdomen ventrally greyish-white with light brown epigastric fold, light grey square area in the posterior part of abdomen, no dark ring around spinnerets. Five retrola- teral cheliceral teeth, three prolateral chelicer- al teeth. Legs: Leg formula presumably 1-4- 2-3, but portions of the first legs missing. Pa- tella-tibia III shorter than IV. Prolateral sur- face of femur I brown, shiny, its upper part and remaining parts of the segment with in- conspicuous, sparse, short whitish setae. Palp: Figures 45-48. — Comparison of left palpal tibia dorsally in four species of Bavia. 45, Bavia aeri- ceps\ 46, Bavia sexpunctata', 47, Bavia fedor new species; 48, Bavia sonsorol new species. Pedipalps light brown, almost bare, prolateral surface of palpal tibia, broad, with anterior edge curved and rising dorsally; spots of white setae on dorsal half of patella and dor- sally on cymbium, just in front of tibia, with triangular apophysis. Female: Female is unknown. Material examined. — Only the holotype. Distribution. — Known only from the is- land of Sonsorol in the Caroline Islands. Genus Cosmophasis Simon 1901 Discussion. — Approximately 40 species are currently placed in this genus by Proszyhski (1990), but he also states that the [nine] Af- rican species are not congeneric with those from Asia. No species of the genus have been reported previously from the Pacific Island re- gion (as here defined). Bonnet (1956), quoted by Proszyhski (1990), mistakenly listed C. maculiventris Strand 1911, from Indonesia, as occuiTing in Polynesia. Two species from New Guinea are illustrated by Chrysanthus (1968). Diagnosis. — Medium-sized unidentate sal- ticids of Simon’s group Chrysilleae. Distin- guished from other Pacific genera of the group by; ocular quadrangle parallel-sided or wid- ening posteriad, long embolus and septate epi- gynum (Figs. 50, 56), and the covering of flat- tened iridescent scales over much of the body. Descriptive notes. — Cephalothorax rela- tively low (42-50% of length of cephalotho- 122 THE JOURNAL OF ARACHNOLOGY Figures 49-57. — Cosmophasis arborea new species from Yap in the Caroline Islands. 49, General appearance of male; 50, Left palp ventrally; 51, Palp laterally; 52, Tibial apophysis dorsally; 53-55, Variation in tibial apophysis of another specimen — ventrally, laterally, and dorsally; 56, Epigynum; 57, Internal structure of epigynum, single spennatheca and ducts. rax), with the highest point at the level of eyes III, from which it very gently slopes anteriorly and posteriorly to the posterior thoracic slope. When alive iridescent due to dense, minute scales which could be seen on preserved spec- imens only with difficulty. Abdomen elongate (with parallel sides in males, slightly more oval in females), gradually tapering posteri- Map 4. — Distribution of three new species of Cosmophasis in the Pacific. Cosmophasis arborea new species (★), Cosmophasis lami new species (□), and Cosmophasis muralis new species (A). orly; spinnerets cylindrical and relatively long. Frontal aspect: eyes I aligned straight along dorsal-most point of their rim, ALE di- ameter of AME, clypeus very low, in some cases obsolete, chelicerae of average size, length about 1 Vi diameters of AME. Ped- ipalps slender. One retrolateral cheliceral tooth, two prolateral cheliceral teeth. Legs: Thin and long, tibia I has two ventral rows of three spines each and two prolateral spines. Cosmophasis arborea new species Figs. 49-57; Map 4 Holotype. — Male from Yap, Yap Island, Colonia, litter and shaking trees on hill behind Protestant mission near bay, 31 May 1973 (J.W. BeiTy) (BPBM). Etymology. — The name refers to the fact that many of the specimens were taken from trees. Discussion. — The species is closely related to Cosmophasis marxi (Thorell 1890a) (Pro- szynski 1984:21-23) from Java. More distant relations include Cosmophasis laticlavia (Thorell 1892) from Sumatra, Cosmophasis BERRY ET AL.— PACIFIC SALTICIDS 123 olorina (Simon 1901b) from Sri Lanka and the species from Java misidentified in the Simon collection as C '‘thalassina’\ All these species are relatives of the type species C. thalassina (C.L. Koch 1846) as redescribed by Zabka 1988, which differs by the anterior lo- cation of embolus, which arises from a broad basis, separated from the bulb, and by longer tibial apophysis. Diagnosis. — Origin of embolus at about midlength of bulb prolaterally, shape of tibial apophysis and cymbium of male palp, and form of epigynum and its internal structure distinguish arborea from other Pacific species of the genus. Description. — Male: {n — 5). Total length 4. 7-5.4 (x = 5.10), length of carapace 2. 1-2.3 (x = 2.21), maximum carapace width 1.5-1. 9 (x = 1.75), eye field length 1.0-1. 7 (x = 1.24), eye row I width 1.3-1. 5 (x = 1.40). Cephalothorax brown, eye field darker brown, lower rim of cephalothorax dark brown to al- most black. Abdomen dark dorsally, with in- distinct whitish, median lines and traces of 2- 3 transverse lines; margins of dorsal surface with whitish rim, posterior tip of abdomen and spinnerets black. Minute dark bristles scat- tered over abdomen. Frontal aspect yellow, framed by dark brown eye field and blackish edge of clypeus; eyes I suiTounded with white and yellowish scales; clypeus darker with a few large, colorless scales. Chelicerae fawn, near rectangular, apically depressed, with higher retrolateral and apical rim; the apical edge ends in a triangular tooth-like process, from that point the edge turns diagonally and is armed with two conical teeth, opposite fang’s tip. Pedipalps slender, white with darker brown-yellow cymbium and prominent bent tibial apophysis. Palpal cymbium mod- erately elongate, with embolus arising anteri- orly from a broad base, separated from bulb, tibial apophysis longer than related species, in form of a slightly twisted plate, with a diag- onally cut tip; there is some variation in the shape of apophysis (Figs. 52-55). Ventral as- pect: mouth parts light brown, sternum yellow with greyish-brown margin, coxae and ventral surfaces of femora yellowish-white; abdomen ventrally greyish-brown. Legs: Leg formula 1 -4-3-2, patella-tibia III shorter than IV. Pa- tella-tibia I length 2. 1-2.6 (x = 2.38). Legs thin and long, yellow, legs I slightly darker with grey spot laterally near extremities of tib- ia. Female: (n = 3). Total length 4.4-5. 1 (x = 4.85), length of carapace 2.05-2.10 (x = 2.07), maximum carapace width 1.45-1.55 (x = 1.50), eye field length 1.05-1.10 (x = 1.08), eye row I width 1.1 -1.4 (x = 1.27). Cepha- lothorax light brownish-yellow with slightly darker brown eye field; covered with ad- pressed scales, practically invisible, except on darker areas around eyes; lower rim of ceph- alothorax brown. Abdomen light brownish- yellow, with indistinct thin irregular lines of darker scales, one scale broad, along lateral margins, and three transverse lines; spinnerets greyish-yellow. Minute dark bristles scattered over abdomen. Frontal aspect yellow, framed dorsally by brown eye field; eyes I sunounded with white and yellowish scales; clypeus ob- solete; with a triangle of longer scales be- tween AME. Chelicerae yellow. Pedipalps white, with yellow tarsus. Ventral aspect whit- ish-yellow. Legs: Leg formula 4- 1-3-2, patel- la-tibia III shorter than IV. Patella-tibia I length 1.3, 1.5 (n = 2). Legs I thin and long, yellow. Epigynum: With two nanow grooves separated by a narrow rise, edges of plate rounded posteriorly (Figs. 56, 57). Material examined. — CAROLINE ISLANDS: Yap, Colonia, night lighting, 1 ?, 26 February 1980 (JWB). Colonia, litter and tree shaking on hills be- hind Evangelical Mission, Id, 31 May 1973 (JAB & JWB). Colonia, St. Mary’s school, sweeping bushes, 19, 1 March 1980 (JWB). Dinay village, Pandanus/grass sweeping, 2dl92imm, 4 March 1980 (JWB). Gitam, shrub shaking. Id, 8 April 1980 (JAB & JWB). Gilman, shaking mango tree, 1 d limm, 15 April 1980 (JAB & JWB). Gagil-Tom- il, shaking bananas, 2d4imm, 2 May 1973 (JWB & JAB). Colonia, tower hill, shaking, ld5imm, 29 May 1973 (JWB & JAB). Map, Chool, tree shak- ing, 7d2imm, 12 April 1980 (JWB & JAB). Distribution. — Known only from Yap in the Caroline Islands. Cosmophasis lami new species Figs. 58-60; Map 4 Holotype. — Male from Fiji, Viti Levu, Suva, Lami beach, on shrub foliage, 3 May 1987. (J.A. Beatty & E.R. Berry) (BPBM). Etymology. — The specific name is a noun in apposition after the locality where the spec- imens were collected. Diagnosis. — The long embolus originating 124 THE JOURNAL OF ARACHNOLOGY Figures 58-60. — Cosmophasis lami new species from Fiji. 58, Left palp ventrally; 59, Palp laterally; 60, Tibial apophysis dorsally. at the retrolateral comer of the bulb and three- lobed palpal tibial apophysis distinguish Cos- mophasis lami from males of other Pacific species. Description. — Male: (n = 2). Total length 5.3, 5.6; length of carapace 2.3, 2.3; maximum carapace width 1.5, 1.5; eye field length 1.1, 1.1; eye row I width 1.4, 1.5. Cephalothorax uniform orange except for dark eye field; cov- ered with colorless to slightly brownish scales, except eye field which has dense orange setae, elongate scales above rims of eyes I whitish. Abdomen ventrally lighter with large colorless scales. Dorsal surface orange, with lighter yel- low marginal line, in one specimen there are also two indistinct lines of brown scales just above marginal line, in the mid-length of ab- domen; posterior tip of abdomen slightly darker, spinnerets blackish-grey. Minute dark bristles scattered over abdomen, almost invis- ible. Frontal aspect: yellowish-fawn; eyes I sunounded with whitish scales; clypeus ap- pears bare but has minute sparse brown setae and bristles. Chelicerae orange, apically light yellow. Legs: Leg formula 1 -4-3-2, patella- tibia III shorter than IV. Patella-tibia I length 1.7, 1.7. Legs I deep yellow, other legs vari- ous shades of yellow. Palp: With rounded bulb, embolus arising at the five o’clock po- sition and half-encircling bulb medially. Tibial apophysis with three triangular rami. Female: Female is unknown. Material examined. — FIJI: Viti Levu, Suva, Lami Beach, on shrub foliage, the holotype and 1 other d, 3 May 1987 (ERB & JAB). Distribution. — Known only from Viti Levu, Fiji. Cosmophasis muralis new species Figs. 61-65; Map 4 Holotype. — Male from Koror, Palau, on lab building, 8 March 1973 (J.A. Beatty & J.W. Beny) (BPBM). Etymology. — Muralis, of the walls, be- cause several specimens were collected on the outer walls of buildings. Diagnosis. — The nearly straight embolus, about half the length of the cymbium, arising from anterior margin of the bulb, shape of male palpal tibial apophysis (resembling that of Cosmophasis chlorophthalma (Simon 1 898)) and form of the epigynum and its outer ducts separate C. muralis from the other Pa- cific species of the genus. Description. — Male: {n = 5). Total length 6. 1-7.7 (x = 6.86), length of carapace 2. 5-3. 2 (x = 2.82), maximum carapace width 1.9-2. 3 (x = 2.01), eye field length 1.2-1. 4 (x = 1.30), eye row I width 1.6-1. 9 (x = 1.69). Cephalothorax brown with a transverse line of whitish scales behind eye field, eye field darker brown with colorless broad scales; ven- Figures 61-65. — Cosmophasis muralis new species from Palau in the Caroline Islands. 61, Left palp ventrally; 62, Tibial apophysis dorsally; 63, Palp laterally; 64, Epigynum; 65, Internal stmcture of epi- gynum — single spermatheca and ducts. BERRY ET AL.— PACIFIC SALTICIDS 125 tral margin with streak of whitish scales, bending on face and coming to the sides of AME. Abdomen reddish-brown with thin me- dian whitish streak, and white margin around % of abdomen, ending by small triangular spot; spinnerets darker. Frontal aspect: a row of horizontal bristles above eyes I, which are also surrounded dorsally by dense row of yel- lowish, inconspicuous setae, laterally and ven- trally by whitish setae, clypeus yellowish- fawn, almost bare; chelicerae longer than usu- al, brown, with anterior surface flattened and apically depressed and limited by apical scler- otized ridge. Legs: Relative leg length 1-4- 2 = 3, with patella- tibia III shorter than IV. Pa- tella-tibia I length 2. 3-3. 2 (x = 2.59). Legs light yellow, legs I with darkened prolateral surfaces of femur, tibia and metatarsus. Palp: Whitish, slender (see diagnosis and Figs. 61- 63). Female: (n = 5). Total length 6.3-7. 1 (x = 6.79), length of carapace 2. 6-2. 9 (x = 2.74), maximum carapace width 1. 8-2.1 (x = 1.98), eye field length 1.3-1. 4 (x = 1.30), eye row I width 1.6-1. 7 (x = 1.67). Differs from male by lack of white line behind eye field, mar- ginal white belt developed but indistinct on the background of yellowish lower sides; ab- domen with two transverse white lines in the posterior half, anterior half bisected by thin median line and with thin white margin. Legs: Leg formula 4-3- 1-2, patella-tibia III shorter than IV. Patella-tibia I length 1.7-1. 9 (x = 1.83). Epigynum: With narrow septum and two broad lunate posterior lobes. Material examined.— CAROLINE ISLANDS: Palau, Koror, on entomology lab building, l(3(ho- lotype)! 9 limm, 8 March 1973 (JAB & JWB). Ko- ror, entomology lab on outside building wall, 16, 1 March 1973 (JAB). Koror, entomology laboratory building, 264\mm, 24 February 1973 (JWB & JAB). Koror, taro patch, 194imm, 7 March 1973 (JWB). Koror, Japanese temple ruins, tree shaking, 19, 17 March 1973 (JAB & JWB). Koror, in cave entrance, 1 9, 17 March 1973 (JWB & JAB). Koror, scrub forest in vacant lot, tree shaking, 19,13 Feb- ruary 1973 (JWB & JAB). Koror, entomology lab building, 13, 25 March 1973 (JWB & JAB). Koror, scrub forest in a vacant lot, tree shaking, 132imm, 14 May 1973 (JAB). Babelthuap, Ngaremlengui, in woods, 1319 limm, 21 April 1973 (JWB & JWB). Angaur, banana-betel palm, 23, 27 April 1973 (JWB & JAB). Peleliu, mixed tropical forest, 332916imm, 22 March 1973 (JWB & ERB). Distribution. — Known only from the Palau group of the Caroline Islands. Genus Flacillula Strand 1932 {Flacilla Simon 1901 — preoccupied) Discussion. — The genus contains at present four species described from the Oriental Re- gion and Pacific Islands. Palpal organs of the specimens studied correspond to that of the type species Flacillula lubrica (Simon 1901) (c/. Proszynski 1984:77) from Sri Lanka, gen- eral features of the epigynum agree with that of Flacillula incognita Zabka 1985 from Vi- etnam. Flacilla kraussi Marples 1964 has the typical stridulating apparatus of Pseudicius Simon 1885 and is here transferred to that ge- nus (NEW COMBINATION). Diagnosis. — Small, elongate unidentate sal- ticids with all legs completely spineless ex- cept for one small distal prolateral spine on metatarsus I in both sexes. First legs robust with distal segments very short (tibia length = patella; metatarsus and tarsus shorter). Descriptive notes. — Minute jumping spi- ders with shiny cephalothorax. The body is low and narrow. Abdomen elongate oval, 25- 38% longer than cephalothorax, as broad as cephalothorax; in males covered by scutum, in females covered by thin tegument, with grey pigmented pattern. Face very low, its whole height occupied by AME, clypeus ob- solete, diameter of ALE % of AME, eyes I aligned in a straight line by their dorsal rims. Legs I are robust in both sexes, much broader than the others, with tibia and patella block- like with prominent angles; in male legs I are the longest, while in female the longest are IV (115-120%), legs II and III are practically of the same length in both sexes. The palpal or- gan is very simple; epigynum, also rather sim- ple, so small that details are visible only with a compound microscope. Flacillula minuta (Berland 1929) Figs. 66-68, 71-74; Map 5 Flacilla minuta Berland 1929a. Flacillula minuta (Berland): Strand 1932. Holotype. — Female from Samoa, Upolu, Malololelei, 2000 ft., (Buxton & Hopkins, in BMNH, examined). Description. — Male: {n = 5). Total length 2.7-2. 8 (x = 2.70), length of carapace 1.05- 1.4 (x = 1.24), maximum carapace width 0.7- 0.9 (x = 0.81), eye field length 0.55-0.60 (x 126 THE JOURNAL OF ARACHNOLOGY Map 5. — Distribution of two species of Flacillula in the Pacific. Flacillula minuta (■), Flacillula ni- tens new species (□). = 0.59), eye row I width 0.6-0. 7 (x = 0.68). Cephalothorax shiny brown, with eye field darker, covered with minute papillae, and lines of similar papillae radiating from the fovea over the thorax and sides. Microscopic col- orless setae scattered over eye field. Abdomen with distinct shiny brown scutum, integument with pattern consisting of median grey area with lateral branches, but sometimes merging into larger grey areas. One retrolateral chelic- eral tooth, three prolateral cheliceral teeth, of which two are very small. Legs: Leg formula 1 -4-2-3; patella-tibia III shorter than IV. Pa- tella-tibia I length 0.7-0.95 (x = 0.85). Legs I light brown, remaining legs yellow; width of tibia I 54% of femur I, 61% of the length of respective segment. Palp: Embolus very thin and short, apical on bulb (Figs. 73, 74). Female: {n = 5). Total length 2.8-3.15 (x = 3.01), length of carapace 1.15-1.30 (x = 1.24), maximum carapace width 0.8-0. 9 (x = 0.86), eye field length 1.15-1.30 (x = 1.24), eye row I width 0.7-0.75 (x = 0.72). Cepha- lothorax with more setae than male. Abdomen similar to male. One retrolateral cheliceral tooth, three prolateral cheliceral teeth. Legs: Leg formula 4- 1-3-2; patella-tibia III shorter than IV. Patella-tibia I length 0.65-0.70 (x = 0.68). Leg color as in male; width of tibia I 57% of femur I, 67% of the length of respec- tive segment. Epigynum: With wide central hood (Figs. 67, 68). Figures 66-74. — The genus Flacillula. 66-68, Flacillula minuta. 66, General appearance of male; 67, Epigynum; 68, Internal structure of epigynum — single spermatheca and ducts; 69, 70, Flacillula nitens new species from Ponape. 69, Epigynum; 70, Internal structure of epigynum — single spermatheca and ducts. 71-74, Flacillula minuta. 71, Leg I of male, prolaterally; 72, Abdominal pattern of female; 73, Left palp ventrally; 74, Palp laterally. BERRY ET AL.— PACIFIC SALTICIDS 127 Material examined. — COOK ISLANDS: Raro- tonga, Muri, shaking trees in yard, 1 d 1 $ limm, 25 March 1987 (JWB & JAB). Turangi Valley, shaking trees, elev. 20 m, Ic?, 1 April 1987 (JWB & ERB). Turangi stream valley, near Ngatangia, shaking trees, l(5l93imm, 18 March 1987 (JWB, ERB & JAB). Oneroa Island, sweeping, 19,21 March 1987 (JWB & ERB). Arorangi Village, on citrus trees, 294imm, 18 March 1987 (JWB & ERB). Taakoka Island, shaking trees, 3 92imm, 23 March 1987 (JWB & ERB). CAROLINE ISLANDS: Truk, Moen, coconut trash. Id, 12 June 1973 (JWB & JAB). Distribution. — Known from Caroline Is- lands, Niue, Samoa, and Cook Islands. Flacillula nitens new species Figs. 69, 70; Map 5 Holotype. — Female from Ponape, moun- tain top, tree shaking, 6 June 1973 (J.W. Berry & J.A. Beatty) (BPBM). Etymology. — The name nitens, shining, re- fers to the shiny cuticle of the spider. Diagnosis. — Epigynum narrower than in Flacillula minuta (Figs. 67, 69). Description. — Female: {n = 2). Total length 2.75, 2.80, length of carapace 1.1, 1.1, maximum carapace width 0.7, 0.7, eye field I length 0.50, 0.55, eye row I width 0.6, 0.6. Differs from Flacillula minuta (Fig. 72) by having bare, shiny cephalothorax, abdomen in one specimen is bare and shiny, on second specimen there is grey pigmentation consist- ing of dense, irregular grey spots on light background. One retrolateral cheliceral tooth, three prolateral cheliceral teeth. Legs: Relative leg length 4-1 =2 = 3; patella-tibia III shorter than IV. Patella-tibia I length 0.55 (x = 0.55). Epigynum: With narrow hood (Fig. 69). Male: The male is unknown. Material examined.— CAROLINE ISLANDS: Ponape, mountain top, tree shaking, 19,6 June 1973 (JWB & JAB). East of Kolonia, breadfruit/ ivory nut forest, hand collecting, 1 9 limm, 8 June 1973 (JWB & JAB). Distribution. — Known only from Ponape in the Caroline Islands. Genus Frigga C.L. Koch Diagnosis. — Medium-to-large unidentate salticids distinguished by Galiano (1979) in her revision of the genus by the presence of a bifid apophysis on the male palp and a deep notch, in which lies a scape, in the posterior margin of the epigynum. A South American genus apparently introduced in the Pacific. Frigga crocuta (Taczanowski 1878) Amycus crocutus Tacz. 1878 Sandalodes calvus Simon 1902 Phiale bispinosa Banks 1930 Frigga crocuta (Tacz.): Galiano 1979 Discussion. — This species was formerly in- cluded in the genus Sandalodes Keyserling 1883 as S. calvus Simon 1902 and is appar- ently of South American origin (Galiano 1979), It occurs across the Pacific from South America to Australia. We have collected it only from the eastern Pacific islands, and it is especially common in the Marquesas Islands. It has been illustrated recently by Davies & Zabka (1989). The other eight species of the genus are restricted to South America (Galia- no 1979). Measurements. — Male: {n = 5). Total length 6.4-9. 2 (x = 7.76), length of carapace 3. 2- 4. 6 (x = 3.78), maximum carapace width 2.3- 3.5 (x = 2.84), eye field length 1.4-1. 8 (x = 1.54), eye row I width 1.8-2. 3 (x = 2.02). One retrolateral cheliceral tooth, two prolateral cheliceral teeth (sometimes low and set very close together, appearing to be a sin- gle notched tooth). Legs: Leg formula 1-3 =4-2; patella-tibia III equal to IV. Patella- tibia I length 3. 2-5. 5 (x = 4.14). Female: (n = 5). Total length 6. 2-7. 6 (x = 6.92) , length of carapace 2. 8-3. 2 (x = 3.02), maximum carapace width 1. 9-2.4 (x = 2.20), eye field length 1.2-1. 4 (x = 1.26), eye row I width 1.6-1. 8 (x = 1.74). One retrolateral cheliceral tooth, two prolateral cheliceral teeth. Legs: Leg formula 4-3- 1-2; patella-tibia III=IV. Patella-tibia I length 1. 7-2.0 (x = 1.92) . Material examined. — MARQUESAS IS- LANDS: Nuku Hiva, Taiohae, grassy knoll, elev. 200 m, 2(32920imm, 21 January 1987 (JWB & ERB). Taiohae, Governor’s residence, shaking dead shrubbery, lc39imm, 21 January 1987 (JWB & ERB). Taiohae, trash pile in culvert, l(32 93imm, 21 January 1987 (JWB & ERB). Taiohae, sweeping bushes, l(3l95imm, 22 January 1987 (JWB & ERB). Taiohae, tree shaking, open field, Sd 16imm, 22 January 1987 (JWB & ERB). Taiohae, hibiscus litter, l(33imm, 22 January 1987 (JWB & ERB). Taiohae, sweeping, elev. 800 m, 5imm, 23 January 1987 (JWB & ERB). Taiohae, on buildings, 4imm, 25 January 1987 (JWB & ERB). Taipivai valley, tree shaking, I95imm, 27 January 1987 (JWB & 128 THE JOURNAL OF ARACHNOLOGY ERB). Toovii, tree shaking, 600 m, 1(529 13imm, 27 January 1987 (JWB & ERB). Toovii, tree shak- ing, 600 m, 196imm, 28 January 1987 (JWB & ERB). Cove west of Taiohae, litter near sea, 1 9, 30 January 1987 (JWB & ERB). Airport, almost desert conditions, sweeping low vegetation, 254924imm, 14 February 1987 (JWB & ERB). Hiva Oa, Han- amenu, tree shaking, 153 9 23imm, 4 Febmary 1987 (JWB & ERB). Hanamenu, top of E ridge, 153 98imm, 5 February 1987 (JWB & ERB). Han- amenu, in grass clump, limm, 6 February 1987 (JWB & ERB). Atuona, shaking low vegetation, 2510imm, 8 February 1987 (JWB & ERB). Atuo- na, roadside vegetation, limm, 9 February 1987 (JWB & ERB). Atuona, coconut/philodendron, sweeping and shaking, 15 limm, 10 February 1987 (JWB & ERB). Atuona, shaking roadside vegeta- tion, limm, 10 February 1987 (JWB & ERB). Atu- ona, roadside sweeping, 5imm, 11 February 1987 (JWB & ERB). Fatu Hiva, Hanavave, coconut for- est, sweeping & shaking, 19, 13 February 1987 (JWB & ERB). Ua Pou Is., Hakahatau Airport, 100 m, 29, 12 July 1988 (S.L. Montgomery) (BPBM). Nuku Hiva, Tene Deserte, Ha’atuatua V, 15,2 July 1988 (S.L. Montgomery) (BPBM). TUAMOTU ISLANDS: Rangiroa Is., Avatoru, swept at light, 19, 28 August 1988 (S.L. Montgomery) (BPBM). SOCIETY ISL.: Huahine, Fare, 0—100 m, 15, March 1979 (N.L.H. Krauss) (BPBM). Huahine, Fare, 0-100 m, 15, March 1972 (N.L.H. Krauss) (BPBM). Moorea, Paopao, sweeping grass, 3 imm, 11 January 1987 (JWB & ERB). Tahiti, Tautira, 29, January 1961 (N.L.H. Kiauss) (BPBM). Raia- tea, Uturoa, 19, Februaiy 1961 (N.L.H. Krauss) (BPBM). NEW CALEDONIA: Col de la Pirogue, 19, 14 February 1963 (Yoshimoto) (BPBM). COOK ISLANDS: Rarotonga, Arorangi reservoir, roadside sweeping, elev. 50 m, 2imm, 1 March 1987 (JWB & ERB). Arorangi village, grass litter, limm, 4 March 1987 (JWB & ERB). Arorangi village, tree shaking, limm, 11 March 1987 (JWB & ERB). Arorangi village, tree shaking, elev. 30 m, 152 94imm, 14 March 1987 (JWB, ERB & JAB). Arorangi village, tree shaking, near Raemam, elev. 250 m, limm, 14 March 1987 (JWB, ERB & JAB). Turangi village, tree shaking, elev. 100 m, 1 9 limm, 1 April 1987 (JWB, ERB & JAB). Inland from Muri Beach, tree shaking, elev. 20 m, limm, 4 March 1987 (JWB & ERB). Rutaki Road, tree shaking, 15 limm, 4 March 1987 (JWB & ERB). Tuoro hill, elev. 200 m, tree shaking, 453imm, 10 March 1987 (JWB & ERB). Matavera, 0-100 m, 1 9, March 1976 (N.L.H. Krauss) (BPBM). Avarua, 0-100 m, 153 93imm, February 1979 (N.L.H. Kiauss) (BPBM). Avarua, 0-100 m, 19, December 1977 (N.L.H. Krauss) (BPBM). Avarua valley, 0-150 m, 29, November 1977 (N.L.H. Krauss) (BPBM). Distribution. — South America, Australia, New Caledonia and the Cook, Galapagos, Mangareva, Marquesas, Society and Tuamotu Islands. Genus Ligurra Simon 1903 Diagnosis. — Small fissidentate salticids with sternum narrow anteriorly and first coxae close together. General appearance short, broad and short-legged, similar to Rhene Tho- rell 1869 and Stertinius Simon 1890. Differs from Rhene by having simpler genitalia in both sexes, from Stertinius by having third eye row not at posterior edge of flat part of carapace, by lacking lateral spines on meta- tarsus I, and retrolateral cheliceral tooth bi- cuspid rather than tricuspid. The latter char- acter may be unreliable. Only one other species is placed in the genus at present (see diagnosis of the new species below). Descriptive notes. — Fissidentate, with bi- cusp tooth. Male chelicerae concave with a ridge edging the concavity; fang with two an- terior spurs which close to either side of a distal flap on the paturon; female chelicerae normal, no spurs, flap or concavity. Ligurra opelli new species Figs. 75-80 Holotype. — Male, mixed tropical forest, woods below SDA school, Airai, Babelthuap, Palau, Caroline Islands, 1 1 March 1973 (Ber- ry, BeiTy & Beatty) (BPBM). Etymology. — This species is named for Dr. Brent Opell, an arachnologist at Virginia Polytechnic Institute and State University at Blacksburg, Virginia. Diagnosis. — Male with embolus tapering, but almost straight, that of L. latidens (Doles- chall 1859) is distinctly curved. Tibial apoph- ysis of palp slender and straight. Female with epigynal openings located near a narrow me- dian hood. The female of L. latidens has no hood and the openings are at the posterior edge of the epigynum (Figs. 75-80). Description. — Male: {n = 5). Total length 2. 9-3. 3 (x = 3.13), length of carapace 1.3- 1.7 (x = 1.52), maximum carapace width 1.3- 1.5 (x = 1.46), eye field length 0.9-1. 1 (x = 0.96), eye row I width 1.1-1. 3 (x = 1.23). Stocky build, flattened, with abdomen over- hanging sloping part of the thorax. Cephalo- thorax yellowish-brown, covered with sparse, whitish setae, ventral edge dark brown and BERRY ET AL.— PACIEIC SALTICIDS 129 Figures 75-80. — Ligurra opelli new species from Caroline Islands. 75, General appearance of male; 76, Left palp ventrally; 77, Palp laterally; 78, Tibial apophysis dorsally; 79, Epigynum; 80, Inter- nal structure of epigynum showing single sperma- theca and ducts. bare, with a narrow line of short white setae above. Lateral eyes naiTowly surrounded with black pigmentation, eyes II in the middle be- tween ALE and III. Abdomen yellowish- brown with indistinct pattern of lighter spots; the entire dorsum covered by scutum. Face low and broad, eyes I in a straight line, ALE separated from AME by about half their di- ameter, which diameter is half that of AME, clypeus reduced, light brown with a thin line of white setae under ALE only. Setae encir- cling dorsal rims of eyes I are white in male, contrasting with darker clypeus. Chelicerae short and broad, anteriorly concave, with re- trolateral edge sclerotized; cheliceral fang with two anterior triangular protuberances and corresponding flap on prolateral edge of che- licera near fang: these are developed to dif- ferent degrees in males of the same sample — from prominent to nearly invisible. One retro- lateral cheliceral tooth, three prolateral cheli- ceral teeth. Ventral aspect: light brown to yel- low, with abdomen lighter. Legs: Leg formula 1 -4-2-3; legs I distinctly longer than others. Legs short, the first very robust, II-IV less so. Patella-tibia I length 1.1 -1.7 (x = 1.44) with patella-tibia III shorter than IV. Palp: See di- agnosis and Figs. 76-78. Female: {n = 5). Total length 3. 6-4. 7 (x = 3.92), length of carapace 1.6-1. 8 (x = 1.69), maximum carapace width 1.5-1. 7 (x = 1.58), eye field length 0.9-1. 1 (x = 1.06), eye row I width 1.3-1. 4 (x = 1.35). Setae encircling dorsal rims of eyes I are colorless in female; also, entire clypeus covered with long white setae. Abdomen, except for the margins, cov- ered by scutum. One bicusp retrolateral chel- iceral tooth, three prolateral cheliceral teeth. Legs: Leg formula 1 =4-2-3; patella-tibia III shorter than IV. Patella-tibia I length 1.1 -1.5 (x = 1.26). Epigynum: See diagnosis and Figs. 79, 80. Material examined. — CAROLINE ISLANDS: Palau Islands, Kayangel Atoll, coconut/R(3r- ringtonia, tree shaking, 2329, 22 May 1973 (JWB). Kayangel Atoll, in cycad tree, 13, 22 May 1973 (JWB). Koror, scrub forest in vacant lot, tree shaking, 533 910imm, 13 March 1973 (JWB & JAB). Koror, scrub forest in vacant lot, tree shaking, 3329l9imm, 13 February 1973 (JWB). Peleliu, tree shaking, Casuarina forest, 137imm, 21 March 1973 (JWB & ERB). Pulo Anna, tree shaking, co- conut/shrub, 13,7 April 1973 (JWB & ERB). Son- sorol, mixed tropical forest, 1 $7imm, 6 April 1973 (JWB & ERB). Arakabesan, mixed tropical forest, elev. 50-75 ft., tree shaking, 134imm, 16 February 1973 (JWB & ERB). Babelthuap, Airai, woods be- low SDA school, mixed tropical forest, tree shak- ing, 13, 11 March 1973 (JAB & JWB). Malakal, dry tropical forest, tree shaking, 1 $ , 14 March 1973 (JWB, ERB & JAB). Ponape, Kolonia, on building, 19, 28 March 1980 (JAB). Distribution. — Known from the Caroline Islands: Palau District and Ponape. Genus Plexippus C.L. Koch 1846 Diagnosis. — A medium-to-large cosmo- tropical unidentate genus belonging to Simon’s Plexippeae. Proszynski (1990) lists 56 species in the genus. Differs from other genera of that group by having two whorls of spines (rather than three) on metatarsus III, by the broad angular bulb of the male palp and the epigynal structure, which lacks the large “windows” and narrow septum of Palpelius Simon 1903 (figs. 109, 112). One retrolateral cheliceral tooth, two prolateral cheliceral teeth. 130 THE JOURNAL OF ARACHNOLOGY Map 6. — Overlapping distribution of two species of Plexippus in the Pacific. Plexippus paykullii (A), a widely distributed pantropical species, and Plex- ippus petersii (A), which is also known from India and Mozambique. Plexippus paykullii (Audouin 1825) Map 6 Anus paykullii Aud. 1825 Plexippus paykulli (Aud.): Paesi 1883 Apamamia bocki Roewer 1944 NEW SYNONYMY Discussion. — This widespread cosmo- tropical species is synanthropic and is consid- ered as a “tramp” species. It has been illus- trated many times, recently by Proszynski (1987) and Zabka (1990). Bonnet (1958) lists numerous other synonyms. Measurements. — Male: {n = 5). Total length 7. 5-8. 9 (x = 8.46), length of carapace 3. 5-4. 5 (x = 4.09), maximum carapace width 2. 9-3. 2 (x = 3.08), eye field length 1.7-1. 9 (x = 1.82), eye row I width 2. 2-2. 5 (x = 2.35). Legs: Leg formula 4- 1-3-2; patella-tibia III shorter than IV. Patella-tibia I length 3.6- 4.3 (X = 4.00). Female: {n = 5). Total length 8.1-10.1 (x = 9.14), length of cai-apace 3.4-4. 3 (x = 3.84), maximum carapace width 2.5-3. 1 (x = 2.76), eye field length 1.6-1. 8 (x = 1.75), eye row I width 2. 1-2.5 (x = 2.23). Legs: Leg formula 4-3- 1-2; patella-tibia III equal to IV. Patella-tibia I length 2.5-3. 1 (x = 2.76). Material examined. — MARSHALL ISLANDS: Eniwetok, 5(3209 30imm (JWB & JAB). HAWAII: Midway Is., Id295imm (J. Richardson). MALAY- SIA: Penang, 13 (JAB). MARQUESAS IS- LANDS: Nuku Hiva, 437 9 12imm (JWB & ERB). Hiva Oa, limm (JWB & ERB). FIJI: Viti Levu, 2319 (JAB, JWB & ERB). SOCIETY ISLANDS: Moorea, 13 (JWB & ERB). TUAMOTU IS- LANDS: Manihi, 5 95imm (ERB). Distribution. — Pantropical species, widely distributed; cosmopolitan in warm climates, overlaps distribution of Plexippus petersi on some islands. N & S America (Mexico, SE USA: to Texas); Mediten'anean (including Is- rael), S Europe, Africa, S & E Asia, Australia, Oceania. Plexippus petersii (Karsch 1878) Map 6 Euophrys petersii Karsch 1878 Plexippus petersi (Karsch): Simon 1903 Discussion. — This species, less common than Plexippus paykullii, overlaps the distri- bution of that species in Fiji and in Majuro (Marshall Islands). However, P. paykullii, al- though often taken on buildings, is frequently found associated with forests or other vege- tation, while P. petersii is more strictly limited to buildings. Almost all of the specimens re- ported here were found associated with build- ings (except for those from uninhabited Helen Reef). Illustrated by Zabka (1985) and Pro- szynski (1987). Measurements. — Male: {n = 5). Total length 5. 6-7. 3 (x = 6.34 ), length of carapace 2. 9-3. 5 (x = 3.14), maximum carapace width 2. 0-2. 7 (x = 2.40), eye field length 1.3-1. 6 (x = 1.44), eye row I width 1.8-2. 1 (x = 1.98). Legs: Leg formula 4- 1-3-2; patella-tibia III shorter than IV. Patella-tibia I length 2.3- 3.2 (X = 2.70). Female: {n = 5). Total length 6. 5-9. 9 (x = 7.90), length of carapace 3. 0-3. 8 (x = 3.38), maximum carapace width 2. 2-2. 7 (x = 2.44), eye field length 1.3- 1.7 (x = 1.52), eye row I width 2. 0-2. 3 (x = 2.14). Legs: Leg formula 4-3- 1-2; patella-tibia III shorter than IV Pa- tella-tibia I length 2. 0-2. 6 (x = 2.36). Material examined. — MARSHALL ISLANDS: Majuro, 13 19 limm (JWB & JAB). CAROLINE ISLANDS: Palau, Angaur, 1 9 (JWB & ERB); He- len Reef, 13 19 limm (JWB & ERB); Koror, 83109 limm (JWB, ERB & JAB). Peleliu, 1 9 limm (JWB & ERB); Malakal, 29 (JWB, ERB & JAB). Yap, 133 9 limm (JWB, ERB & JAB). Truk, 19 (JWB & JAB). Ponape, 2 9 limm (JWB & JAB). FIJI: Viti Levu, 1329 (JWB, ERB & JAB). AMERICAN SAMOA: Tutuila, 1319 (JAB). AUSTRALIA: Darwin, 19 (JAB). Distribution. — Mozambique, India, Sri Lanka, Singapore, Japan, China, New Guinea, Solomon Islands, Caroline Islands, Marshall Islands, Fiji, Samoa, Australia. BERRY ET AL.— PACIFIC SALTICIDS 131 Figures 81-90. — The genus Thorelliola. 81-85, Thorelliola ensifera from the Marquesas Islands. 81, Left palp ventrally; 82, Palp laterally; 83, Epigynum; 84, Internal structure of epigynum — single sper- matheca and ducts. 85, Abdominal pattern of female. 86-90, Thorelliola dumicola new species, from Ponape. 86, Left palp ventrally; 87, Palp laterally; 88, Epigynum; 89, Internal stmcture of epigynum — single spermatheca and ducts. 90, Abdominal pattern of female. Genus Thorelliola Strand 1942 {Thorellia Keyserling 1882 preoccupied) Diagnosis. — A fissidentate salticid with one 3-6 cusped retromarginal cheliceral tooth, two prolateral cheliceral teeth, one much smaller than the other. Metatarsi I with lateral spines, coxae one widely separated, anterior eye row strongly recurved. Thorelliola ensifera (Thorell 1877) Figs. 81-85 Plexippus ensifer Thorell 1 877 Thorellia ensifera (Thorell): Keyserling 1882. Discussion. — There is some evidence (Chelstowska pers. comm.) that the wide- spread Pacihc salticid that has always gone by this name is not the same species as Plexippus ensifer, which was originally described from Celebes. We have taken it in a variety of hab- itats: in litter, on tree trunks, foliage and build- ings, in forested and non-forested areas, and at elevations from 0-800 m. Measurements. — Male: {n = 5). Total length 4.4-4.8 (x = 4.68), length of carapace 2. 1-2.3 (x = 2.19), maximum carapace width 1.6-1. 8 (x = 1.73), eye held length 1.1-1. 3 (x = 1.21), eye row I width 1.5-1. 7 (x = 1.67). Legs: Leg formula 1-4-2-3; patella-tibia III equal to IV. Patella-tibia I length 1.9-2. 2 (X - 2.02). Female: (n = 5). Total length 3. 7-4. 3 (x = 3.92), length of carapace 1.7-1. 8 (x = 1.74), maximum carapace width 1.3-1. 9 (x = 1.45), eye held length 0. 9-1.0 (x = 0.97), eye row I width 1.3-1. 9 (x = 1.44). Legs: Leg formula 4- 1-3-2; patella-tibia III equal to IV. Patella- tibia I length 1.1-1. 3 (x = 1.16). Material examined. — (all from our collection) MARIANA ISLANDS: Guam, 633 9, 7imm. CAROLINE ISLANDS: Palau, 343319, 75imm; Yap, 213399, 59imm; Ulithi, 2339, limm; Pon- ape, 1349, 6imm. MARSHALL ISLANDS: Eni- wetok, 16315 9, 3 limm; Kwajalein, 15359, 12imm; Majuro, 6317 9, 28imm. FIJI: Viti Levu, 283279, 56imm. AMERICAN SAMOA: Tutuila, 635 9, 17imm. COOK ISLANDS: Aitutaki, 5389, 13imm. Rarotonga, 383369, 97imm. SO- CIETY ISLANDS: Moorea, 21315 9, 85imm. MARQUESAS ISLANDS: Fatu Hiva, 839 9, 19imm; Hiva Oa, 273389, 65imm; Nuku Hiva, 22317 9, 41imm. Distribution. — Malaysia across the Pacihc to the Marquesas Islands. 132 THE JOURNAL OF ARACHNOLOGY Thorelliola dumicola new species Figs. 86-90 Holotype. — Male from Ponape (Caroline Islands), SW of Sekere Sch., shaken from bushes overhanging roadbank, 10 June 1973. Coll. J.A. Beatty & J.W. Beiry. Etymology. — The specific epithet, dumic- ola, means dwelling in thickets, because of the habitat in which the specimens were collected. Discussion. — The placement of this species in the genus Thorelliola can be questioned be- cause of its lack of the two strong recurved frontal spines characteristic of T. ensifera. However, the color pattern, while paler and less distinct in dumicola, is almost identical with that of light-colored and juvenile speci- mens of T. ensifera. Body shape and append- age proportions are similar in both species, and the genitalia are of the same form. The number and aiTangement of spines on the ap- pendages show only slight differences be- tween the two species, and both have the un- usual 3-6 cusped tooth on the cheliceral margin. Diagnosis. — Differs from T. ensifera by its paler abdominal pattern, and in males by hav- ing only a single slender frontal bristle rather than two strong ones; recognizable by palp and epigynum (Figs. 86-89). Description. — Male: {n = 1). Total length 2.9, length of carapace 1.5, maximum cara- pace width 1.2, eye field length 0.8, eye row I width 1.2. Cephalothorax greyish-brown with somewhat lighter eye field and anterior, flat thorax; eye field rather bare, with small indistinct setae. Abdomen very different from the usual coloration of T. ensifera, resembling rather Euophrys, light whitish-yellow with in- distinct pattern of yellowish-grey pigmented median and marginal spots, making indistinct chevrons in posterior half; dark bristles and setae more prominent. Frontal aspect: face brownish, eyes I suiTounded with whitish se- tae, bristles on clypeus small and inconspic- uous; chelicerae brownish; pedipalps whitish with yellow cymbium. Legs: Leg formula 4-3- 1=2; patella-tibia III equal to IV. Patella- tibia I length 0.9. Legs I light yellow. Palp: Lacks the enlarged spines and somewhat angular cymbium in T. ensifera, but in form of palpal bulb and tibial apophysis the two are extreme- ly similar (Figs. 86, 87). Female: (n = 2). Total length 3.1, 4.0; length of carapace 1.5, 1.9; maximum cara- pace width 1.3, 2.0; eye field length 1.0, 1.1; eye row I width 1.3, 2.0. Cephalothorax yel- low with brown shade, with a pair of brown spots on posterior slope of thorax. Small, col- orless setae poorly visible on light back- ground, however sparse, short dark bristles more distinct because of color contrast. Ab- domen coloration like male (Fig. 90). Frontal aspect: face light yellow, eyes I surrounded with light yellow setae dorsally, ventrally whitish; chelicerae bulging basally, greyish- yellow, apically yellow. Legs: Relative leg length 4-3- 1-2; patella-tibia III equal to IV. Patella-tibia I length 1.1, 1.4. Pedipalps and legs I yellowish-white with sparse dark setae. Epigynum: Of same form as that of T. ensif- era, but with spermathecae smaller and ducts shorter (Figs. 83, 84; 88, 89). Material examined. — CAROLINE ISLANDS; Ponape, SW of Sekere Sch., shaken from bushes along roadbank, 10 June 1973, Id (the holotype) and l9limm (JAB & JWB) (BPBM). Mt. top, shaking, 6 June 1973, 19 (JAB & JWB). Distribution. — Known only from Ponape in the Caroline Islands. Genus Trite Simon 1885 Discussion. — Related species that have been redescribed by Zabka (1988) are Trite auricoma (Urquhart 1885), T. pennata Simon 1885, and T. planiceps Simon 1889. Seven- teen species are currently placed in the genus (Proszynski 1990), all from Australia, New Zealand and south Pacific islands. Four of these, T. rapaensis Berland 1942 (Figs. 91- 93), T. ignipilosa Berland 1924 (Figs. 94, 95), T. lineata Simon 1885 (Figs. 98-100), and T. gracilipalpis Berland 1929 (Figs. 101-104) are illustrated here for comparison with the new species and to make available more de- tailed illustrations than have been available before. Diagnosis. — Medium-to-large fissident or plurident salticids with fourth leg longer than third, cephalic portion of the cephalothorax flat, second row of eyes closer to anterior than posterior row, anterior eyes nearly contiguous and ocular area wider behind than in front. Trite ponapensis new species Figs. 96, 97; Map 7 Holotype. — Male from Ponape, top of mountain, tree shaking, 6 June 1973 (J.W. Beny & J.A. Beatty) (BPBM). BERRY ET AL.— PACIFIC SALTICIDS 133 Figures 91-104. — The genus Trite. 91-93, Trite rapaensis’, 91, Left palp ventrally; 92, Palp laterally; 93, Epigynum. 94, 95, Trite ignipilosa from New Caledonia; 94, Left palp ventrally; 95, Palp laterally. 96, 97, Trite ponapensis new species from Ponape; 96, Left palp ventrally; 97, Palp laterally. 98-100, Trite lineata from Noumea. 98, Epigynum; 99, Tibia I, female; 100, Ventral view of female chelicera; 101-104, Trite gracilipalpis . 101, Left palp ventrally; 102, Palp laterally; 103, Ventral view of chelicera; 104, Dorsal appearance of male (TYPE) from Loyalty Island. Etymology. — ^The species is named for the island of Ponape in the Caroline Islands, the only known location. Discussion. — This species is placed in Trite with considerable doubt. It is unidentate rather than fissidentate and the middle eye row is closer to the posterior than the anterior eyes. The general appearance is similar to Bavia, but both the cheliceral teeth and genitalia are quite different from that genus. In some re- spects it resembles Thiania C.L. Koch 1846 and the Marpissa C.L. Koch 1846 group of genera. With only a single specimen available we are unable to arrive at any firm conclusion. Diagnosis. — Externally resembles Bavia Simon 1877 from which it differs by the struc- ture of the palp, the straight dorsal edge of the palpal tibia and the cheliceral dentition. Dis- tinguishable from other Trite by the long thin, nearly straight embolus which originates near the base of the bulb (Figs. 96, 97). Description. — Male: (n = 1). Total length 134 THE JOURNAL OF ARACHNOLOGY Map 7. — Distribution of five species of Trite in the Pacific. Trite ponapensis new species (□), Trite gracilipalpis (■), Trite ignipilosa (A), Trite lineata (A), and Trite rapaensis (★). 9.5, length of carapace 4.1, maximum cara- pace width 3.0, eye field length 2.1, eye row I width 2.5. Cephalothorax broad but more elongate and less swollen than in Bavia; light brown, dorsum of thorax with broad lighter zone with minute whitish setae, fovea promi- nent. Eye field light greyish-brown, finely rough, slightly shiny, with a line of whitish setae along lateral eyes and above eyes I. Sides with sparse, inconspicuous brownish se- tae, longer and denser under lateral eyes along a crest reminiscent of Ascyltus. Abdomen long and thin, broadest anteriorly, gradually nar- rowing and pointed posteriorly, with long spinnerets. Median area of dorsum whitish, margins brown, sides with dense thin brown lines on white background; spinnerets light yellow. Frontal aspect-face light reddish- brown, rims of eyes I black surrounded with inconspicuous white-tipped setae. No con- trasting spots. A line of long brown setae be- low AME overhanging cheliceral bases, che- licerae reddish-brown with papillate surface. ALE almost touching AME, diameter of AME = 2 diameters of ALE. One retrolateral chel- iceral tooth, two prolateral cheliceral teeth. Ventral aspect: Endites elongate, rectangular, rounded apically, nan'owing basally, without any depression or separate expansion along external edge. Chelicerae, endites, and labium fawn, anterior coxae light brown, coxae II-IV whitish-yellow; sternum yellow, darker mar- ginally; abdomen ventrally greyish- white with light brown epigastric fold, light grey rectan- gular area along majority of length of abdo- men, no dark ring around spinnerets. Pedi- palps with dense brush of long grey setae along edges of patella, tibia and cymbium, no such characters in any Bavia. Pedipalps long and thin, yellowish-brown, extremities of seg- ments slightly lighter, but no contrasting spots. Dorsal anterior edge of pedipalpal tibia tri- angular. Legs: Leg formula 1 -4-2-3, patella- tibia III shorter than IV. Patella-tibia I length 4.2. Legs I light brown, in some areas yellow- ish; femur I with prolateral surfaces bald and shiny, brown, with a crest of blackish setae along dorsal edge. Inconspicuous spot of whit- ish setae prolaterally on patella. Tibia I has one additional lateral spine between first and second ventral spines, slightly more dorsally, with all ventral prolateral spines concentrated in the apical half of the segment. Palp: See diagnosis and Figs. 96, 97. Female: Female is unknown. Material examined. — Only the holotype. Distribution. — Known only from Ponape in the Caroline Islands. ACKNOWLEDGMENTS We are especially grateful for the Academic Research Grants from Butler University to one of the authors (JWB) which helped sup- port the field work and enabled one of the authors (IP) to work on this project in the US. The US Department of Energy provided travel funds for the work at Eniwetok and Kwajalein in the Marshall Islands. Two travel grants from the Indiana Academy of Science to one of the authors (JWB) helped support this work. A grant (#PB 0442/P2/93/04) from the Committee for Scientific Research in Poland helped support the work of one of the authors (JP). Elizabeth Ramsey BeiTy’s contributions to all phases of the field work in the Pacific and at home have been invaluable. We are grateful to the staff of the Bishop Museum, Honolulu for the loan of specimens and for assistance in many ways and to the American Museum of Natural History for the loan of specimens. We also wish to thank the staff of the Richard Gump Laboratory, Moorea, So- ciety Islands; Dr. Kamlesh Kumar at the For- estry Station, Tholo-I-Suva, Dr. Madhu Ka- math and Mr. Satya Ram Singh at the Koronivia Research Station in Fiji; Ozanne Rohi in Hiva Oa and the Forestry Department in Nuku Hiva (Marquesas Islands), Rick Wel- land in Rarotonga (Cook Islands), and Josie BERRY ET AL.— PACIFIC SALTICIDS 135 and David Sadaraka, Aitutaki (Cook Islands). Also, Sakie Morris, Demei Otobed and Rubak Obak in the Palau Islands provided valuable assistance. In the Yap Islands, Mel Lundgren, Gabriel Ayin, Sister Ann Dowling and Margie Falanruw contributed greatly to our work. Without their cooperation our field work would have been much less pleasant and ef- fective. LITERATURE CITED Audouin, V. 1825. Explication sommaire des planches d'Arachnides de PEgypte. . . In Savig- ny, Description de PEgypte. . . Histoire Naturel- le, Vol. 1. Paris. Pp. 1-339. Berland, L. 1924. Araignees de la Nouvelle Cale- donie et des iles Loyalty. In Sarazin, E & J. Roux. Nova Caledonie, ZooL, 3:159-255. Berland, L. 1929a. Araignees (Araneida). In In- sects of Samoa and other Samoan terrestrial ar- thropods, Part viii, 2:35-78. London. Berland, L. 1929b. Araignees recueillies par Ma- dame Pruvot aux iles Loyalty. Bull. Soc. Zool. France, 54:388-399. Berland, L. 1934a. Les Araignees du Pacifique. Mem. Soc. Biogeo., 4:155-180. Berland, L. 1934b. Araignees de Polynesie. Ann. Soc. Ent. France, 103:321-336. Berland, L. 1938. Araignees de Nouvelles Hebri- des. Ann. Soc. Ent. France, 107:121-190. Berland, L. 1942. Polynesian Spiders. Occ. Papers B. P. Bishop Museum, 17:1-24. Berry, J.W., J.A. Beatty, & J. Proszyhski. 1996. Salticidae of the Pacific Islands. L Distribution of twelve genera, with descriptions of eighteen new species. J. ArachnoL, 24:214-253. Bonnet, P. 1956. Bibliographia Araneorum, 2: 1242. Toulouse. Bonnet, P. 1958. Bibliographia Araneorum, 2: 3717. Toulouse. Chrysanthus, Fr. 1968. Spiders from South New Guinea. X. Tijdschrift voor Entomologie, 111: 49-74. Davies, V.T & M. Zabka. 1989. Illustrated keys to the genera of jumping spiders (Araneae: Saltici- dae) in Australia. Mem. Queensland Mus., 27: 189-266. Doleschall, C. 1859. Tweede Bijdrage tot de Ken- nis der Arachniden van den Indischen Archipel. Act. Soc. Sci. Ind.-NeerL, 5:1-60. Galiano, M.E. 1979. Revision del genero Frigga C. L. Koch, 1851. (Araneae, Salticidae). Acta Zool. Lilloana, 33:113-135. Karsch, F. 1878. Diagnoses Attoidarum aliquot no- varum Novae Hollandiae collectionis Musei Zoologici Berolinensis. Mitt. Munchen Ent. Ver., 2:22-32. Koch, C.L. 1846. Die Arachniden, Dreizehnter Band. Niirnberg. Pp. 1-104. Koch, C.L. 1851, Ubersicht des Arachnidensys- tems. Heft 5. Niirnberg. Pp. 1-104. Koch, L. 1865. Beschreibungen neuer Arachniden und Myriopoden. Verb, zool.-bot. Ges. Wien, 15: 857-892. Marples, B.J. 1955a, Spiders from some Pacific is- lands, Pacific Science, 9:69-76. Marples, B.J. 1955b. Spiders from western Samoa. J. Linn. Soc. Zool., 42:452-504. Marples, B.J. 1957. Spiders from some Pacific is- lands. II. Pacific Science, 11:386-395. Marples, B.J. 1964. Spiders from some Pacific is- lands. Part V. Pacific Science, 18:399-410. Proszynski, J. 1984. Atlas rysunkow diagnosty- cznych mniej znanych Salticidae. Zesz. Naukowe WSRP, Siedlce. Part 1. 177 pp. Proszynski, J. 1987. Atlas rysunkow diagnosty- cznych mniej znanych Salticidae. Zesz. Naukowe WSRP, Siedlce. Part 2. 172 pp. Proszynski, J. 1990. Catalogue of Salticidae (Ara- neae), a synthesis of data since 1758. Wyzsza Szkola Rolniczo-Pedagogiczna w Siedlcach. Sie- dlce. 366 pp. [updated versions available on computer disc and on the Internet at http://spi- ders.arizona.edu/proszynski/proszynski.html] Simon, E. 1877. Etudes arachnologiques. 5e Me- moire. ix. Arachnides recueillis aux iles Philip- pines par MM. G. A. Baer et Langlaise. Ann. Soc. Ent. France, 5:53-96. Simon, E. 1885. Materiaux pour servir a la faune arachnologique de la Nouvelle Caledonie. Ann. Soc. Ent. Beige, C.R., 29:87-92. Simon, E. 1900. Arachnida. In Fauna Hawaiien- sis. . . London. Simon, E. 1901a. Histoire Naturelle des Araignees. Tome 2, fasc. 3. Paris. Pp. 381-668. Simon, E. 1901b. Descriptions d’ Arachnides nou- veaux de la famille des Salticidae (suite). Ann. Soc. Ent. Beige, 45:141-161. Simon, E. 1902. Descriptions d’especes nouvelles de la famille Salticidae (suite) Ann. Soc. Ent. France, 71; 389-421. Strand., E. 1932. Miscellanea nomenclatoria zool- ogica et paleontologica, III. Folia Zool, Hydro- bioL, 4:133-147. Strand, E. 1942. Miscellanea nomenclatoria zool- ogica et paleontologica, X. Folia Zool. Hydro- bioL, 11:386-402. Taczanowski, L. 1878. Les Araneides du Perou. Famille des Attides. Bull. Soc. Imp. Nat, Mos- cou, 53:278-374. Thorell, T. 1877. Studi sui Ragni Males! e Papuan!. Part I. Ragni di Selebes raccolti nel 1874 dal Dott. O. Beccari. Ann. Mus. Civ. Stor. Nat. Gen- ova, 10:341-634. Thorell, T 1881. Studi sui Ragni Males! e Papuan!. 136 THE JOURNAL OF ARACHNOLOGY Part III. Ann. Mus. Civ. Stor. Nat. Genova, Pp. 1-720. Thorell, T. 1890a. Studi sui Ragni Malesi e Papu- an!. Part IV, 1. Ann. Mus. Civ. Stor. Nat. Genova, 8:1-419. Thorell, T. 1890b. Aracnidi di Nias e di Sumatra raccolti nel 1886 dal Sig. E. Modigliani. Ann. Mus. Civ. Stor. Nat. Genova, 10:5-106. Thorell, T. 1892. Studi sui Ragni Malesi e Papuan!. Part IV. Ann. Mus. Civ. Stor. Nat. Genova, 31: 1-490. Wanless, E 1978. A revision of the spider genus Sobasina (Araneae: Salticidae). Bull. British Mus. Nat. Hist. (ZooL), 33:245-257. Zabka, M. 1985. Systematic and zoogeographic study on the family Salticidae (Araneae) from Viet-Nam. Ann. Zool. Warszawa, 39:1-485. Zabka, M. 1987a. Salticidae (Araneae) of Oriental, Australian and Pacific regions. I. Ann. Zool. Warszawa, 40:437-450. Zabka, M. 1987b. Salticidae (Araneae) of Oriental, Australian and Pacific regions. II. Genera Lyci- das and Maratus. Ann. Zool. Warszawa, 40:451- 482. Zabka, M. 1988. Salticidae (Araneae) of Oriental, Australian and Pacific regions. III. Ann. Zool. Warszawa, 41:421-478. Zabka, M. 1990. Salticidae (Araneae) of Oriental, Australian and Pacific regions. IV. Genus Ocri- siona Simon 1901. Rec. Australian Mus., 42:27- 43. Zabka, M. 1991. Salticidae (Araneae) of Oriental, Australian and Pacific regions. V. Genus Holo- platys Simon, 1901. Rec. Australian Mus., 43: 171-240. Zabka, M. 1991. Salticidae (Araneae) of Oriental, Australian and Pacific regions. VI (“VII”)- Mop- solodes, Abracadabrella and Pseudosynagelides -new genera from Australia. Mem. Queensland Mus., 30:621-644. Zabka, M. 1992. Salticidae (Araneae) of Oriental, Australian and Pacific regions. VII. Grayenulla and Paraplatoides -new genera from Australia and Caledonia. Rec. Australian Mus., 44:165- 183. Zabka, M. 1992. Salticidae (Araneae) of Oriental, Australian and Pacific regions. VIII. A new ge- nus from Australia. Rec. Western Australian Mus., 15:673-684. Zabka, M. 1993. Salticidae (Araneae) of Oriental, Australian and Pacific regions. IX. Genera Af- raflacilla Berland & Millot 1941 and Evarcha Simon 1902. Invert. Taxon., 77:279-295. Zabka, M. 1994. Salticidae (Arachnida: Araneae) of Oriental, Australian and Pacific regions. X. Genus Simaetha Thorell. Rec. Western Austra- lian Mus., 16:499-534. Zabka, M. 1995. Salticidae (Arachnida: Araneae) of Oriental, Australian and Pacific regions. XL A new genus of Astieae from Western Australia. Rec. Western Australian Mus. Supp., 52:159- 164. Manuscript received 8 September 1995, revised 23 August 1996. 1997. The Journal of Arachnology 25:137-176 THERAPHOSIDAE OF THE MOJAVE DESERT WEST AND NORTH OF THE COLORADO RIVER (ARANEAE, MYGALOMORPHAE, THERAPHOSIDAE) Thomas R. Prentice: Department of Entomology, University of California, Riverside California 92521 USA ABSTRACT. Two new species of Aphonopelma from the Mojave Desert are described, A. joshua and A. mojave. Four nominal species, Aphonopelma iodium (A. iodius), A. melanium (A. melanius), A. angusi, A. nevadanum, described from the Mojave Desert and the Great Basin are treated as a single species; Aphonopelma iodium (Chamberlin 1939) is proposed as the species name since it is one of two possible senior synonyms (the other, A. melanium (A. melanius) in the same publication) and is the specific name of the first Theraphosidae to be described from the Mojave Desert; A. iodium is redescribed. Generic, subgeneric, and specific characters previously used to separate Aphonopelma are reviewed. Aphonopelma is redefined; Clavopelma is removed from the synonymy of Aphonopelma. The status of the following eight species described prior to 1939 with type localities either in the United States or in Baja, California is discussed: Aphonopelma californica, A. leiogaster (Doleschall), A. steindachneri (Ausserer), A. rusticum, A. marxi, A. helluo (Simon), A. rileyi (Marx), and A. pseudoroseum (Strand). Aphonopelma steindachneri, A. rusticum, A. marxi, and A. helluo are considered as valid species; A. californica, A. leiogaster, A. rileyi, and A. pseudoroseum are considered as nomina dubia. The Aphonopelma of North American are poorly known. Although many species have been described few specimens can be properly identified either by using available keys or by wading through species descriptions. Most identifiable specimens belong to species found in Mexico or Central America that are easily recognized by unique color patterns, such as that of A. seemanni. Correct identification of specimens collected within the United States is often suspect since determinations must be based on the process of elimination using cob lection dates and locality data in combination with coloration, coxal setation, and metatarsal scopulation. Chamberlin & Ivie (1939), Chamberlin (1940) and Smith (1994) described the major- ity of the currently recognized Aphonopelma species. Since many of their descriptions were based on one or two specimens variational limits were unknown to them with the con- sequential result of species determination by highly variable, artificial characters. It is my intention in this paper to describe the thera- phosid species of the Mojave Desert using re- liable taxonomic characters, established as such only after thorough analyses of variation within each putative species. Ecological, geo- graphical, and behavior data are included and support morphologically-based species deter- minations. The species redescribed here, A. iodium, be- longs to a species group that are virtually im- possible to differentiate by unaided visual ex- amination: all are very similar in size, color, and extent of metatarsal scopulation and all share a common fall breeding season (breed- ing season determination was based on the time of year that both type males and males from type localities were collected). I will re- fer to the group collectively as 'eutylenum types’ or as the " eutylenum group’ since A. eutylenum, as member of this assemblage, is referred to in the literature more than any of the other species and since preliminary data suggest that many of these species (other than those considered A. iodium) belong to a single widely distributed species, in which case A. eutylenum would be considered the senior subjective synonym based both on page pri- ority and usage in the literature. METHODS All specimens analyzed in this study were mature individuals. Males of the different spe- cies were collected while searching for fe- 137 138 THE JOURNAL OF ARACHNOLOGY males during their respective breeding sea- sons. Breeding season is defined as the period of time between which the first males of a given species abandon their burrows, becom- ing itinerants, and the vast majority have died from senescence or predation to the point that individuals are rarely found until the follow- ing year. The majority of males collected were taken between the years 1989-1993; and bor- rowed specimens other than types were col- lected as early as 1966. The system for col- lection was to drive slowly down least traveled backroads or powerline roads on a weekly basis (often 2-3 days a week) starting 1-2 weeks before the perceived beginning of a particular breeding season and ending when no males were found for two consecutive weeks. The east and west Mojave Desert are de- fined as east and west, respectively, of a north- south line through Death Valley, Silurian Val- ley, Silver Dry Lake, Soda Dry Lake, the Bristol Mountains, Devil’s Playground, Bristol Dry Lake, Cadiz Valley, and Danby Dry Lake. Joshua Tree National Park and Death Valley National Park are, respectively, refeiTed to as Joshua Tree National Monument and Death Valley National Monument in this manuscript because they were so-named while this study was in progress. Measurements were performed using an American Optical 570 stereomicroscope equipped with an eyepiece micrometer. Mea- surements are all in millimeters and are ac- curate to 0.05 mm except for tarsal measure- ments which are accurate to 0. 1 mm. Leg and pedipalp measurements were taken from the left side unless some or all segments of a giv- en leg were missing or it was apparent that an appendage was in the process of regeneration. All segment measurements were performed from the retrolateral aspect; this measurement equal to the distance from the proximal point of articulation to the distal most point of the segment (Coyle 1971, 1989). Carapace and sternum lengths were taken with anterior and posterior margins in the same horizontal plane; width measurements were performed in the same fashion. Measurements of femur width in males were taken from the dorsal as- pect at the widest pro- to retrolateral point other than at the base; in femur I & IV that point is preapical of its articulation with the patella; in femur III that point is basad of the preapical point for femur I in the distal half of the segment. Extent of metatarsal scopula- tion was determined by using maximum ex- tent of complete metatarsus I scopula as the proximal point for measurement in metatarsi II-IV. Because tarsal measurements were difficult to perform without removing much of the dis- tal scopulae and claw tufts, measurements were performed by depressing the apical scop- ulae and claw tufts against the integument with a flattened forceps. Because of a greater possibility of eiTor in measuring the tarsi, leg I-IV ratios were calculated by adding the lengths of the femur, patella, tibia, and meta- tarsus only to represent the length of each leg; palp length was calculated by adding the lengths of the femur, patella, and tibia only. Cheliceral length measurements also were dif- ficult to perform because of relaxation and over-extension of chelicerae as a result of preservation. These measurements were omit- ted except in type specimens of A. Joshua new species and A. mojave new species because accurate measurements generally required dis- section of chelicerae and because in living conspecific specimens chelicerae can be rela- tively more distended in some individuals of a smaller size than in other individuals of a larger size. Abdominal measurements were also omitted, except in type specimens and in largest and smallest specimens, due to depen- dence of abdominal size on nutritional state of the specimen. Carapace length was found to be the most reliable indicator of overall size. All ink drawings were done with the aid of a camera lucida fitted on a Wild Heerbrugg M5 stereomicroscope. Spermathecae were cleared in 20% NaOH prior to illustration. Scanning electron micrographs were taken with a JEOL JSM C35 scanning electron mi- croscope. Characters and quantitative character values in Tables 1 and 2 are an essential part of each species description. These tables should be re- ferred to unless specific values are given in the text. Specimens examined. — Specimens examined other than samples of the species described here, are as follows: types: Clavopelma tamaulipecmn, Chaunopelma radinum, Aphonopelma angusi, A. iodium {iodius), A. melanium (melanius), A. neva- danum, A. eutylenum, A. paloma, A. phanus, A. phasmus, A. reversum, A. simulatum, A. zionis PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 139 (AMNH), Eurypelma steindachneri [syntype- BMNH and (2) presumed types-NHMW], E. rusti- cum [cotype #1585, 47(36)-USNM], non-types: E. marxi [45(30), 43(20)-USNM] and E. helluo [50 (44)-USNM], and specimens from the type locali- ties of the following species: A. anax, A. apacheum, A. behlei, A. brunnium (brunnius), A. chalcodes, A. clarum, A. coloradanum (coloradana) A. cratium (cratius), A. cryptethum (cryptethus), A. echinum (echina), A. helluo, A. lithodonum, and A. seeman- ni. Abbreviations used. — Spination abbrevia- tions: a = apical; b = basal; d = dorsal half; e = preapical; Fe = femur; L = left; m = medial; Me = metatarsus; p = prolateral di- rection; Pt = patella; r = retrolateral direction; R = right; Ti = tibia; v = ventral half; 0.25, 0.80, etc. = approximate location of a spine taken as a fraction of the segment length from the proximal end. Tabular character table abbreviations: At- ten = attenuate; bas swln = basally swollen; Cxi = coxa I; E = east Mojave; LAI, LAII, LAIII, LAIV - lengths of legs I, II, III, IV, respectively; LC = carapace length; LFI, LFIII, LFIV, LFP = lengths of femora I, III, IV, palp, respectively; LMI, LMII = lengths of metatarsi I and II, respectively; LTI, LTII, and LTIV = lengths of tibiae I, II, and IV, respectively; par div = partially divided; retr bnd = retrolateral bend; S&Cx == sternal and coxal; ScMIV = scopula of metatarsus IV; ScTalV = scopula of tarsus IV; sht = short; swln = swollen; undiv = undivided; unif = uniform; v == ventral; W = west Mojave; WC = carapace width; WCh = chelicerae width; WFI, WFIII = widths of femora I and III, respectively; WS = sternal width. Museum abbreviations: AMNH = Ameri- can Museum of Natural History, New York; BMNH = The Natural History Museum, Lon- don; MNHN = Museum National d’Histoire Naturelle, Paris; NHMW = Naturhistoisches Museum, Wien; USNM == National Museum of Natural History, Washington. Other abbreviations: BDM = Beaver Dam Mountains, southwestern Utah; JTNM = Joshua Tree National Monument, California. Abbreviations for eyes are standard for Ara- neae. TAXONOMY Synonymy of Rhechostica = Aphonopel- ma. — Raven (1985) placed the following gen- era in the synonymy of Rhechostica Simon 1892, a senior subjective synonym of Apho- nopelma Pocock 1901: Aphonopelma, Duge- siella Pocock 1901, Pterinopelma Pocock 1901, Delopelma Petrunkevitch 1939, Chau- nopelma Chamberlin & Ivie 1939, and Cla- vopelma Chamberlin 1940. He concluded that they shared the form of the double tibial spur and the thorn-like setae on the prolateral cox- ae and that there were no other characters known that merited their continued separation. Because of the extensive usage of Aphono- pelma in the literature Levi & Kraus (1989) petitioned the ICZN to give Aphonopelma precedence over Rhechostica. By Opinion 1637 of the ICZN (June 1991) Aphonopelma was given precedence whenever the two were considered to be synonyms. Review of generic characters. — Pocock (1901) erected six genera during his dismem- berment of Eurypelma, two of which had members north of Mexico, Aphonopelma and Dugesiella. These two genera were respec- tively distinguished by the absence and pres- ence of a plumose scopula on the prolateral surface of femur I and the retrolateral surface of the palpal trochanter and by spiniform and thom-like setae on prolateral coxa I. Petrun- kevitch (1939) erected Delopelma which he differentiated from Dugesiella by the com- plete absence of plumose hairs and from Aphonopelma by the presence of simple, re- cumbent hairs on coxae and trochanters. Chamberlin (1940) apparently recognized the presence of plumose setae in all genera and considered Delopelma (retaining only the type D. marxi) a subgenus of Aphonopelma based on the similar form of setae on prolateral coxa I. He, in turn, erected Chaunopelma which differed from both Aphonopelma and Duge- siella by the presence of fine, soft prone hairs on the anterior coxa and trochanter of leg I and on the posterior palpal trochanter. Raven considered the differences in coxa I setation artificial. Other than by the setation of prolateral coxa I and the form of the double tibial spur. Raven distinguished Aphonopelma {Rhechostica) by the following characters: scopula of tarsus IV integral (no setal division), an area of plumose or spatulate hairs on retrolateral maxillae or palpal trochanter, and males with a slender and tapering embolus. The type species of two of the genera in the synonymy of Aphonopel- 140 THE JOURNAL OF ARACHNOLOGY ma, Clavopelrna and Pterinopelma, are en- dowed with setae on the retrolateral palpal tro- chanter that may be termed spatulate in form. In Pterinopelma, Pocock likened these setae to those found in Euathlus {Brachypelma) which are stout, finely plumose (often long plumed), and lanciform. Although less stout in Clavopelrna these lanciform setae are dis- tinctly different from the relatively slender, plumose, hair-like or spiniform setae (in some species finely plumose) of the remaining Aphonopelma. Smith characterized Aphonopelma by the following: no organs of stridulation or plumose setae/hairs on trochanter or coxa of palp or leg I or on femur I, no plumose pad present on femur IV, a tapering embolus (more stout and shorter in material from Mexico and Central America) with a simple keel on dorsal surface or on apex (sometimes ribbed or toothed), sper- mathecae composed of sepai'ate seminal recep- tacles, integral tarsus IV scopula, scopula of metatai'sus IV ranging from and no swol- len leg segments. One of the character distinc- tions that Smith used to sepai*ate Euathlus {Brachypelma) and Aphonopelma was the pres- ence of plumose setae on the prolateral tro- chanter and basal femur of leg I in Euathlus and the presence of non-plumose setae in Aphonopelma. It is not clear whether Smith was referring to the absence of lanciform setae (believed to have a stridulatory function) in Aphonopelma or had not detected the plumose condition of the hair-like or spinifoiTn setae in the genus. Smith removed Aphonopelma pal- oma from the genus as the type of his newly erected monotypic genus, Apachepelma, based on partial division by setae of the tarsus IV scopula in combination with small size. Of the known Aphonopelma species, only A. Joshua shares this character combination with Apache- pelma paloma although it is usually slightly larger (Table 1). However, both species are moiphologically more similar to A. mojave (eastern race), a species in which the tarsus IV scopula is entire, than to each other. Males of an undescribed Aphonopelma species (females not known) from southeast Arizona appeal* to be, otherwise, very similar to A. paloma males except that in this species the tarsus IV scopula is entire. Aphonopelma crinita is a consider- ably larger species than A. paloma but also shows partial division of the tarsus IV scopula (Perez-Miles 1994). In size, prolateral coxal se- tation, and palpal bulb morphology it is more similar to congeneric species with entire tarsus IV scopula than to the other species with par- tially divided scopulae. Because Apachepelma was erected on characters not recognized as ge- nerically significant and because the type spe- cies shares with Aphonopelma all characters di- agnostic of the genus, Apachepelma is here considered a synonym of Aphonopelma. In ad- dition to males of A. paloma, males of both A. Joshua and eastern A. mojave have the third femur swollen to a degree that is easily rec- ognized and that is non-overlapping with spe- cies in which the third femur is slightly swol- len. The extent of the scopula of metatarsus IV within the genus was found to vary from a few scattered distal hairs (A. paloma) to over 80% scopulate distolaterally. Setation of the retro- lateral femur IV in all Aphonopelma examined was found to be similar or identical to that of the prolateral femur I. Male emboli were found to have two mai'ginal or lateral keels (usually indistinct in males with slender emboli) and one or two prominent medial keels. Although Raven did not consider the form of setae on anterior coxa I to warrant generic division, it does appear to be of value in dis- tinguishing species groups within those Aphonopelma in which males have a slender embolus. For instance, hair-like setae, as in A. radinum, are also characteristic of three sim- ilar small species, A. paloma, A. Joshua (Fig. 7) and A. mojave (Fig. 8); basally swollen spi- niform setae are characteristic of A. iodium (Figs. 9, 10) and most other southwestern spe- cies; thorn-like, apically filiform setae, very distinct from the homologous setae of A. io- dium, are characteristic of most species east of Utah and Arizona. To my knowledge, spec- imens with thorn-like setae have not been col- lected in the United States west of Globe, Ar- izona. Critical review of subgeneric taxonomic characters. — Chamberlin considered Aphon- opelma to consist of three subgenera, Delo- pelma, Gosipelma, and Aphonopelma. Delo- pelma and Gosipelma were differentiated on the basis of the number of spines on the an- terior face of the male palpal tibia (two and four, respectively). Several species within these subgenera were additionally distin- guished by spination of the ventral palpal tibia and of the palpal patella. Smith (1994) also used spination characters in diagnoses of new PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 141 species. For example, he separated A. chain- bersi Smith from A. cryptethum Chamberlin by the presence of four and five prolateral spines, respectively, on the palpal tibia, dif- ferences in the shape of the posterior half of the basal segment of the palpal bulb, and dif- ferences in the shape of the cuspules on the labium. However, in Aphonopelma paloma the numbers of spines on the ventral and prolat- eral palpal tibia and on the palpal patella were shown to be variable and in all species de- scribed herein tibial and patellar spination var- ied more intraspecifically than between or among Chamberlin’s Delopelma and Gosipel- ma species. For example, in A. Joshua new species males (from one locality) the palpal tibiae were armed with 3-6 prolateral and 1- 3 ventral spines and the palpal patellae with 0-2 spines; in A. mojave new species males tibiae were armed with 2-7 prolateral and 0- 4 ventral spines and patellae with 0-2 spines; in A. iodium males (from BDM) tibiae were armed with 4-8 prolateral and 1-3 ventral spines and patellae with 0-3 spines. As a re- sult of the high variability in spination, dis- tinctions between species based on these char- acters are considered artificial differences. Chamberlin separated Aphonopelma from Delopelma and Gosipelma by relative lengths of metatarsus and tibia I; “tibia I longer than metatarsus I” in Delopelma and Gosipelma; “tibia I not longer than metatarsus I, usually clearly shorter” in Aphonopelma. Males of both A. Joshua new species and western A. mojave new species have been found in which the ratio is reversed from the usual condition; in both species tibia I can be longer than, equal to, or shorter than metatarsus I within a single population. These conditions invalidate the Aphonopelma subgenus distinction. Al- though the length ratio of tibia I to metatarsus I has been discarded as a subgeneric character in this study it is taxonomically significant be- cause it distinguishes the southern Utah pop- ulation of A. mojave from other eastern pop- ulations, reveals generalities of reversal between A. Joshua and western A. mojave, and is one of the quantitative characters that dis- tinguishes A. mojave from A. radinum and A. iodium from similar types described prior to 1939. Critical review of specific taxonomic characters. — All ocular characters used by Chamberlin to separate subgroups within the subgenus Aphonopelma were found to be highly variable in this study and are consid- ered to be artificial differences. For instance, A. clarum and A. eutylenum were distin- guished from the other Aphonopelma (subge- nus) species in Chamberlin’s key on the basis of “lateral eyes separated by the diameter of a posterior one or nearly so” versus separated “at most but little more than the radius of the posterior eye”. Lateral eyes of the A. eutyle- num holotype are separated by approximately 0.70 X PLE length. This criterion would place two apparently conspecific Red Mountain, California males in separate species; distance between lateral eyes in one is 0.44 X PLE length and in the other is 0.83 X PLE length. Specimen size and coarseness of setae on the anterior face of coxae were used by Cham- berlin to separate A, clarum and A. eutylenum. The ratio of carapace length of the former spe- cies to the latter is 0.73 which is less of a difference than the ratio of 0.66 for the small- est to largest A. iodium males from the BDM. Although size can be an important character in separating species, range values are needed before its discriminating value can be deter- mined. The spiniform setae on prolateral coxa I in both A. clarum and A. eutylenum are of the same form found in A. iodium (Figs. 9, 10). The coarseness of these setae was found to be relative to specimen size in A. iodium and their dispersion over the prolateral surface varied slightly within a given population. Chamberlin used general color of legs and abdomen to separate A. melanium and A. io- dium. The former holotype, collected in Sep- tember toward the beginning of the breeding season, was described as “gunmetal brown or blackish”, and the latter, collected in late No- vember toward the end of the breeding season, was described as “fighter brown or yellow- ish”. North American tarantulas are darker in color shortly after a molt than they are at any other time prior to a subsequent molt (pers. obs.). A reasonable assumption is that A. me- lanium was darker in color because it was col- lected closer to the time of its definitive molt than A. iodium, which had faded substantially by the time it was collected in late November. Although coloration can be a reliable charac- ter in separating some species, distinction be- tween shades of a particular color can be high- ly subjective because of temporal changes in a specimen’s color. 142 THE JOURNAL OF ARACHNOLOGY Table 1. — Males of Aphonopelma: Taxonomic characters and quantitative character values which sep- arate the Mojave Desert Aphonopelma and distinguish them from the most similar species; ° — species with hairlike setae on prolateral coxa I, distinguished by superscript only from A. joshua{^) and A. mo- javeC). * — included in the synonymy of A. iodium\ ' — no overlap with A. Joshua; ^ — no overlap with A. mojave; ^ — no overlap with A. iodium; [Type] — holotype; USNH — non-types, Marx collection. Mean and standard deviation shown in parentheses (8.33, 0.56). Abbreviations defined in Methods section of text. Carapace measurements are in millimeters. ° Joshua °moJave iodium °paloma marxi simulatum °radinum [n = 25] [n = 42] ro II [n = 9] [2-USNH] (Type) [Type] LC 7.00-9.70 6.70-9.60 9.35-16.90 2'4. 10-6.20 '9.10 "9.85 7.50 (8.33, 0.56) (8.25, 0.72) (13.0, 2.05) (5.33, 0.51) '9.00 LTI/LMl 0.91-1.01 0.92-1.07 0.85-1.01 '1.04-1.11 '2' 1.25 "'1.27 "1.10 (0.97, 0.03) (1.00, 0.04) (0.92, 0.04) (1.08, 0.02) '2' 1.30 LTII/LMII 0.84-0.91 0.86-0.96 0.82-0.91 0.88-0.94 '2' 1.12 "' 1 .09 "0.98 (0.87, 0.02) (0.91, 0.02) (0.87, 0.03) (0.92, 0.02) '2' 1.09 LFI/LTI 1.12-1.19 1.18-1.28 1.17-1.33 '1.22-1.31 2'1.31 •1.22 "1.10 (1.15, 0.02) (1.22, 0.02) (1.24, 0.04) (1.26, 0.03) 2'1.29 LFI/LTII 1.22-1.33 1.30-1.41 1.31-1.41 '1.41-1.53 '2' 1.53 "' 1 .44 21.27 (1.25, 0.03) (1.35, 0.02) (1.36, 0.02) (1.48, 0.04) "-'1.53 LFI/LMI 1.05-1.19 1.15-1.29 1.07-1.24 '1.28-1.40 "'1.63 "‘1.55 ‘1.21 (1.12, 0.03) (1.21, 0.04) (1.15, 0.04) (1.36, 0.04) "•1.67 LFI/LMII 1.05-1.18 1.16-1.31 1.1 1-1.25 '1.28-1.42 "‘1.71 "'1.58 ‘1.25 (1.10, 0.03) (1.22, 0.04) (1.18, 0.04) (1.36, 0.04) "' 1 .67 LAI/LC 3.06-3.58 2.87-3.36 2.95-3.46 2.79-3.24 "‘2.82 "'2.81 3.19 (3.39, 0.1 1) (3.07, 0.1 1) (3.23, 0.11) (3.09, 0.15) "2.87 LAII/LC 2.87-3.40 2.70-3.18 2.73-3.27 2.58-2.98 "'2.52 "'2.64 2.95 (3.25, 0.13) (2.90, 0.10) (3.05, 0.09) (2.83, 0.13) "'2.60 LAIV/LC 3.64-4.15 3.11-3.71 '3.09-3.60 '3.06-3.55 "‘2.88 "'2.94 3.41 (3.89, 0.13) (3.34, 0.12) (3.42, 0.10) (3.36, 0.16) "‘2.96 LAI/LAIV ^0.81-0.89 '0.90-0.95 '0.92-0.96 '0.90-0.95 "‘0.97 "0.96 ■0.93 (0.87, 0.02) (0.92, 0.01) (0.94, 0.01) (0.92, 0.01) "'0.97 LAI/LAIII M. 02- 1.08 '1.09-1.14 '1.10-1.16 2'1.15-1.19 "'1.24 "'1.22 2'1.16 (1.04, 0.01) (1.11, 0.01) (1.13, 0.02) (1.17, 0.01) "‘1.25 LAl/LP -2.28-2.42 '2.04-2.23 '2.05-2.21 '1.96-2.07 "‘1.87 "‘1.85 ‘2.21 (2.35, 0.04) (2.12, 0.05) (2.12, 0.04) (2.02, 0.04) "‘1.90 WFllIAVFI swln > 1.19 swln normal (W) 'normal swln 'normal ‘normal ‘normal normal < 1.15 swln (E) Scopula Mistal distal 2'distal -'distal 'distal ■'distal distal MIV (%) 25-50 25-55 70-85 0-20 40, 45 35 45 Division -par div 'undiv 'undiv -par div ‘undiv ‘undiv 'undiv Sc TalV (%) 25-60 50-100 S & Cx(v) -sht, stout. 'long. 'long. ‘long. 'long. ‘long. ‘long. setae conicform atten atten atten atten atten atten Cx I setae bas unif bas unif ^'bas swln bas unif “‘bas swln ^‘bas swln bas unif prolateral hairlike hairlike spinifm hairlike spinifm spinifm hairlike Palpal Aetr bnd 'retr bnd -retr bnd Aetr bnd Aetr bnd Yetr bnd 'retr bnd bulb uniform angular uniform uniform uniform uniform angular Carapace black black -'paper-bag black chestnut chestnut chestnut color brown and black Smith (1994) described 25 new species of Aphonopelma, 14 of which were described from single specimens, eight from two speci- mens each, and only three from several spec- imens each. Most of these descriptions were based on taxonomic characters that either were determined to be highly variable intra- specifically and widely overlapping interspe- cifically or are subjective in nature. Characters most consistently used in his diagnoses were PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 143 Table 1. — Extended. phasmus zionis ^melanium iodium *angusi *nevadanum helluo rusticum [Type] [Type] [Type] [Type] [Type] [Type] [USNH] [cotype] ■'8.20 '‘1.04 '•0.93 21.17 1.33 ‘1.22 ‘1.24 3.25 3.02 3.55 '‘0.91 2‘1.15 ‘2.09 ‘normal '2‘distal 60 ‘undiv ‘long, atten 2‘bas swln spinifm 2retr bnd angular 2‘light brown 29.70 1.01 0.91 ■'21.15 1.31 1.15 ‘1.19 3.30 3.05 ‘3.56 ‘0.93 2‘1.16 ‘2.15 ‘normal '2‘distal 65 ‘undiv ‘long, atten 2‘bas swln spinifm 2retr bnd uniform 2‘light brown 2‘13.50 0.94 0.86 ‘1.25 1.36 1.17 1.16 3.21 3.06 ‘3.40 ‘0.94 ‘1.13 ‘2.11 ‘normal 2‘distal 72 ‘undiv ‘long, atten 2‘bas swln spinifm 2retr bnd uniform 2‘yellow gray 2‘14.30 0.94 0.89 ‘1.22 1.33 1.15 1.18 23.39 3.20 ‘3.55 2‘0.96 ‘1.14 ‘2.18 ‘normal 2‘distal 79 ‘undiv ‘long, atten 2‘bas swln spinifm 2retr bnd uniform 2‘pale buff 2‘ 10.70 0.97 0.91 ‘1.23 ‘1.38 ‘1.20 ‘1.25 3.29 3.07 ‘3.45 ‘0.95 2‘1.16 ‘2.08 ‘normal 2‘distal 70 ‘undiv ‘long, atten 2‘bas swln spinifm 2retr bnd uniform 2‘yellow gray 2‘ 15.90 2‘0.87 20.85 ‘1.27 1.32 1.10 21.13 3.24 3.12 ‘3.42 ‘0.95 ‘1.11 ‘2.17 ‘normal 2‘distal 71 ‘undiv ‘long, atten 2‘bas swln spinifm 2retr bnd uniform 2‘golden yellow ■'2‘ 17.30 '‘1.03 '‘0.94 ■'2 ‘1.35 '2 ‘1.46 ■'2‘1.38 ■'2‘1.36 ■'2‘2.80 ■'2‘2.67 ■'2 ‘3. 02 ‘0.93 ‘1.13 '2 ‘1.95 ‘normal '2‘distal 60 ‘undiv ‘long, atten 2‘bas swln spinifm 2retr bnd uniform 2‘14.10 0.89 ‘1.41 ‘1.25 ■'2‘2.57 ■'2 ‘2. 96 ‘normal 2‘distal 0.75 ‘undiv 'long, atten 2‘bas swln spinifm 2retr bnd uniform shape of the basal division of the palpal bulb, extent of metatarsus IV scopula, number and/ or position of megaspines of the lower process of the tibial spur, and prolateral spination of the palpal tibia. Of these characters the shape of the basal division of the palpal bulb was weighted most heavily in separating males of new species. In examining the Aphonopelma, including type and non-type males and males collected from the type localities of 12 nom- 144 THE JOURNAL OF ARACHNOLOGY inal species, I found relatively minor variation in the general shape (form) of the basal divi- sion. Intraspecific variation (also see Perez- Miles 1989) was found to be as great as in- terspecific variation in all species described herein and in the following species: A. palo- ma, A. reversum, A. chalcodes, and A. color- adanum (Canon City, Colorado). Conversely, males of three very dissimilar species (A. io- dium, A. reversum, and A. coloradanum) were found in which corresponding divisions of the bulb were nearly identical. Major differences, such as in several illustrations by Smith, may have been the result of molting difficulties, damage to the bulb (through the effects of preservation or during the immediate post- molt sclerotization process), or genetic muta- tion but are only doubtfully indicative of spe- cies distinctions, especially in light of locality data. A second primary character that Smith used to diagnose males was the number and/or po- sition of stout apical or subapical megaspines on the lower process of the male tibial spur. In A. mojave the inner megaspine (preapical spine on the inner or concave surface of pro- cess) was always present while the outer megaspine (spine on the outer or convex sur- face of process) was frequently absent; both spines varied considerably in size, shape, and their position on the process, their articulation varying from almost apical to decidedly preapical. One or more less stout apical spines were occasionally present between the larger megaspines. On the upper process the arma- ture was also found to be variable; the inner surface of the process was always equipped with one megaspine but occasionally with two subequal megaspines and/or one to several ad- ditional lesser spines. A similar degree of variation was also found in both A. Joshua new species and A. iodium. The extent of the scopula of metatarsus IV (weighted heavily by Smith) proved to be a reliable character in this study after range val- ues were established, especially in combina- tion with size and color characters and locality data. Length values from museum specimens were often difficult to determine because much of the metatarsal pubescence and scop- ulae had been worn away through repetitive examination. However, under high magnifi- cation scopular extent could usually be deter- mined by cuticular examination. In species such as A. iodium the lateral extent of the scopula was found to be greater than the me- dial extent but this condition appeared to have been overlooked by Smith in his measure- ments of various type specimens. For the Chamberlin holotypes, A. angusi, A. iodium, A. melanium, and A. nevadanum. Smith illus- trated the approximate extent of metatarsus IV scopula as, the distal 35, 40, 60, and 60% of the segment, respectively. According to Smith’s criteria, this placed the former two species in a different species group than the latter two species. My own measurements of metatarsus IV scopula of the same holotype specimens are as follows (in approximate per- centage): A. angusi, 40 medial, 70 retrolateral; A. iodium, 50 medial, 80 retrolateral; A. me- lanium, 50 medial, 70 retrolateral; A. nevadanum, 60 medial, 70 retrolateral. For the species A. iodium (not including above types) scopula extent ranged from 40-60% medially and from 70-85% retrolaterally (maximum extent). Medial extent was usually greatest in A. Joshua and A. mojave and in both ranged from 25-50%. Other male characters that Smith consid- ered of lesser weight included number of megaspines of the upper process of the tibial spur, prolateral spination of tibia I, shape of the labial and maxillary (used less often) cus- pules, condition of distal embolus (keeled or not), shape of the embolus tip, and shape of the posterior half of the palpal bulb (basal por- tion of the middle division). The number of labial cuspules in A. Joshua varied from 33- 78, in A. mojave from 26-90, and in A. iodium from 70-140 (in specimens collected from one locality). Distribution of the cuspules, hence, shape of the distribution varied consid- erably within all three species with substantial interspecific overlap. Differences in this char- acter are considered to be artificial due to high variability in number and distribution and to the subjective nature in the interpretation of such highly irregular shapes. The shape, orientation, and keeled condi- tion of the embolus have been found to be relatively constant in males with slender em- boli; shape and orientation are essentially as in the palpal bulbs in Figs. 14-21, 29-44 and as described below under the additional di- agnostic characters for Aphonopelma. Even in such diverse species as A. reversum, A. behlei, and A. coloradanum the apical emboli were PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 145 often indistinguishable and the accentuation of the median keel no more variable interspecif- ically than intraspecifically. No males exam- ined in this study were found with significant variation in embolic characters. Rather than attributing the major differences described and illustrated by Smith as characteristic of new species, I would suggest that such conditions were anomalies resulting from genetic abnor- malities, molting problems, or damage to bulbs. If they were representative shapes then such species are indeed rare and distinguished only by the condition of the embolus. The basal portion of the middle division of the palpal bulb was found to be variable with- in all species described here, primarily in the region of the basal cuticular protrusion (prox- imal prolateral protuberance) of the prolatero- dorsal surface. Interspecific variation in this character was found to be greater between some species than intraspecific variation with- in the compared species (Figs. 14-21, odd numbered figures) although was negligible be- tween 'eutylenum types’ (Figs. 29-44, even figures) and between several other species such as A. coloradanum and A. r ever sum. In diagnosing females Smith primarily used two characters, shape of the spermathecae and scopulation of metatarsus IV; other characters used were setation of prolateral coxa I and spination of the palpal tibia. Spermathecae of six A. iodium females (two from one locality and two from a second locality) in Figs. 45- 50 illustrate conspecific variation. Spermathe- cae of two other species, A. joshua and A. mo- jave, are illustrated in Figs. 22-28 to show interspecific similarity as well as intraspecific variation. Clearly, shape of the spermathecae can vary considerably within a species but, conversely, can be very similar among fe- males of different species. Consequently, this character is not considered to have specific discriminating value. Primary taxonomic characters. — In this study the taxonomic characters with the high- est discriminating value were present in both genders; they are: (1) form of setae on the prolateral face of coxa I, (2) extent of meta- tarsal scopulation (primarily, metatarsus IV), (3) condition of tarsal scopula (entire or di- vided), (4) lengths of both legs III and IV rel- ative to length of leg I, (5) lengths of both leg III and IV relative to carapace length, and (6) color of the carapace (in living specimens). Additional characters that were weighted heavily separated males only; they are: (1) lengths of tibia I and metatarsus I relative to each other, (2) lengths of tibia II and metatar- sus II relative to each other, (3) lengths of both tibiae I and II relative to femur I, (4) lengths of both metatarsi I and II relative to femur I, and (5) condition of femur III (swollen or nor- mal). Status of some old ^Eurypelma’ species. — I have included eight Aphonopelma species in this study (seven with type localities in the U.S. and one from Baja California) that were described prior to 1939 (described as Thera- phosa or Eurypelma) because of the possibil- ity of synonymy in name with A. iodium and/ or A. mojave which would preclude the use of one or both as species names for the Mojave Desert tarantulas; they are as follows: Thera- phosa californica, T. leiogaster Doleschall 1852 {Eurypelma, in Ausserer 1871), E. stein- dachneri Ausserer 1875, E. rileyi Marx 1888, E. rusticum, E. marxi, E. helluo Simon 1891, and E. pseudoroseum Strand 1907. Although all of these species are now considered Aphonopelma (consult Raven 1985), I will re- fer to them in this section as Eurypelma. Eurypelma rileyi was described on the basis a single female, type locality, Santa Barbara, California. The type is believed to no longer exist (N.I. Platnick, pers. comm. 1995) al- though Smith (1994) redescribed the species from a specimen in the USNM, maintaining that it was the female holotype. In my ex- amination of the specimen, I found the spi- nation armature (of all legs) to be grossly in- congruent with that described by Marx, leg IV slightly greater than, rather than slightly short- er than, the carapace length (if Marx included coxa IV in his measurement), and no indica- tion on the labeling that the specimen was part of the Marx collection (no data other than that on the label exists). Given these conditions, the likeliness that this specimen is the type of E. rileyi is controvertible. In addition, the epi- gastric region is missing rendering the gender of the fragmented alcohol specimen indeter- minable (the specimen was formerly pinned and dried and the abdomen stuffed with cotton wool). The carapace color of E. pseudoroseum was described as reddish-yellow or pinkish (trans- lation), an obvious condition in living speci- mens but one almost impossible to determine 146 THE JOURNAL OF ARACHNOLOGY in specimens preserved for any length of time. Unfortunately, types of E. pseudoroseum (two females) do not exist and except for carapace color the species description could be equally applicable to a number of species. A male (#1589-BMNH) from the Koch col- lection, type locality Pecos River, Texas, was considered by D.J. Clark (1961) to be the specimen figured by Ausserer in his original description of E. steindachneri (the specimen is fragmented and is missing leg III and IV except for the right trochanter and femur of leg IV; leg II (R) and tibia and metatarsus IV (R) from a larger specimen(s) are mixed in with the type). Smith (1994) also considered this male to be the holotype, assuming that Ausserer described a male from Pecos River, Texas rather than from San Diego, California. In a personal communique that I received from Dr. Jurgen Gruber (NHMW) concerning two male and one female specimens labeled Eurypelma californicum the following infor- mation was conveyed: (1) the original labeling of the three Austrian specimens is feared to have been discarded but museum acquisition records state that two specimens are types of E. steindachneri (California: San Diego) al- though the presumed type series was mixed up with later material and (2) since Ausserer described both genders, the non-type speci- men is suggested to be a male collected in California (circa 1892) which apparently was lumped together with the types. Upon exam- ination of these specimens I discovered that one of the males and the female were, decid- edly, the specimens on which Ausserer based his description of Eurypelma steindachneri. Carapace and leg measurements that I per- formed on both genders were in agreement with those of Ausserer’s, the ocular deformity described of the type female was present in the NHMW female, and the dorsal coverage of the urticating patch and the relatively straight cut and armature of the apical superior tibial spur described of the type male were in congruence with the respective characters of the NHMW male. Contrarily, measurements that I performed on the BMNH specimen were not in agreement with those in Ausserer’s original description. The coloration of the types, as described by Ausserer, is typical of males and faded females, respectively, of specimens I have collected near the Mexican border, just southeast of San Ysidro, Califor- nia. The limited extent of the scopulae of metatarsi III and IV in these San Ysidro spec- imens is very diagnostic of this species (ap- prox.-distal Vi-^A and Vs or less, respectively) and agrees both in character with the NHMW specimens (Ausserer did not include data on the extent of scopulae) and in the general lo- cality data of specimens collected near the Mexican border and provided to Ausserer by Dr. Steindachner. The non-type male is of the same species as the types. Additional information conveyed to me by Dr. Gruber is as follows: (1) of the original specimens of Doleschall, according to Dole- schall, the type of Theraphosa {Eurypelma) californica (female) was a dry specimen and according to Ausserer that of T. {Eurypelma) leiogaster (male) was also a dry specimen and (2) types of these species no longer exist (also verified by Gertsch (1978) in a personal com- munication to W. Icenogle). In the original description of E. rusticum, Simon noted type localities as both Ft. Yuma and Williams, Arizona. It is not entirely clear from which of these localities the described specimen was collected. The holotype and other material from Ft. Yuma are believed to have been lost or destroyed. However, a male in USNM (#1585, cotype, E. rusticum. Col- lection: Marx, type locality Williams, Arizo- na) may be one of the original specimens in Simon’s series (there are also leg segments of a smaller specimen mixed in with the frag- mented type; the male is missing the patella, tibia, metatarsus, and tarsus of leg I). Smith (1994), apparently unaware of the existence of the cotype, redescribed Aphonopelma rusti- cum based on “Simon designated paratype material from northern Mexico’’. Although Simon clearly stated in his description that the species also occun'ed in northern Mexico, Smith selected, as the lectotype, a specimen (#5873, male, MNHP-Paris) from Mazatlan, Mexico, a locality distinctly not in northern Mexico. Since this male is not believed to be of the same species as the cotype male (#1585), the precise identity of E. rusticum = A. rusticum is uncertain. Reexamination of all type material will be necessary before a lec- totype can be objectively designated. The A. rusticum of Chamberlin (type locality, Apache Trail, Arizona) is clearly not the E. rusticum of Simon; Smith redescribed this species as A. rothi. PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 147 In his 1940 work, Chamberlin indicated that the name "marxV had previously been used to cover several different species including A. simulatum (Chamberlin & Ivie 1939) and that the precise identity of the species would re- main in question until either the types were critically restudied or ample material from the San Bernardino Mountains (California) was examined. The labeling of the Marx speci- mens, from which Simon described E. marxi, apparently indicated type localities of Califor- nia: San Bernardino Mountains and New Mexico: Punta-del-Aqua (Marx’ labeling was often suspect (Gertsch 1961)); Chamberlin discounted New Mexico as a type locality of E marxi but, interestingly, considered June Springs, New Mexico a second locality for A. simulatum (male and female). Although no la- beled E. marxi types are known to exist, two E. marxi non-type males in USNM (localities: l-#43(20), California, l-#45(30). New Mexi- co) from the Marx collection match, closely enough, Simon’s original description of the species; the A. simulatum holotype is indistin- guishable from these specimens in all char- acters examined (Table 1). Furthermore, I have spent a considerable amount of time col- lecting in and around the San Bernardino Mountains and have found no specimens ex- hibiting the combination of characters present in the non-type A. marxi and in the A. simu- latum holotype. In light of the above, I sug- gest that Marx mislabeled one of two New Mexico specimens and conclude here that Eu- rypelma marxi = Delopelma marxi = Apho- nopelma marxi is a valid species name, that the New Mexico male be considered the neo- type, and that A. simulatum be considered a junior synonym of A. marxi (NEW SYNON- YMY). E. helluo = Delopelma helluo = A. helluo (Table 1) is also considered here as a valid species name represented by the holotype (#17707, MNHN, type locality. Cape Lucas, Baja California); only one other A. helluo specimen (examined) from the Marx collec- tion (non-type male, USNM, #50(44), locality. Cape Lucas) is known to exist. Nomina dubia. — Because of the inadequa- cy of the original Eurypelma descriptions which could be equally applicable to a number of species combined with the loss of type specimens and/or appropriately labeled non- type specimens from collections from which types were described, Theraphosa californica = E. californica = Dugesiella californica = Aphonopelma californica, E. rileyi = A. rileyi, T. leiogaster = E. leiogaster = A. leiogaster, and E. pseudoroseum = Delopelma pseudo- roseum = A. pseudoroseum should be consid- ered as nomina dubia, “in the interest of pro- moting” nomenclatural stability. Genus Aphonopelma Pocock Rhechostica Simon 1892: 162 (type species by orig- inal designation Homoeomma texense Simon 1891). Suppressed as a senior synonym of Aphonopelma by ICZN Opinion 1637. Aphonopelma Pocock 1901: 553 (type species by original designation Eurypelma seemanni EO. Pickard-Cambridge 1897). First synonymized with Rhechostica by Raven 1985: 149. Dugesiella Pocock 1901: 551 (type species by orig- inal designation D. crinita Pocock 1901). First synonymized with Rhechostica by Raven 1985: 152. Delopelma Petrunkevitch 1939: 567 (type species by original designation Eurypelma marxi Simon 1891). First synonymized with Rhechostica by Raven 1985: 151. Gosipelma Chamberlin 1940: 4 (type species by original designation G. angusi Chamberlin 1940). Originally described as a subgenus of Aphono- pelma-, never elevated to full genus status. First synonymized with Rhechostica by Raven 1985: 153. Chaunopelma Chamberlin 1940: 30 (type species by original designation Delopelma radinum Chamberlin & Ivie 1939). First synonymized with Rhechostica by Raven 1985: 151. Apachepelma Smith 1994: 45 (type species by orig- inal designation Aphonopelma paloma Prentice 1992). NEW SYNONYMY Diagnosis. — The genus Aphonopelma is dis- tinguished from all other genera by the follow- ing combination of characters: (1) no known external organs of stridulation (males do strid- ulate, however); (2) normal, relatively slender (hair-like or spiniform) plumose setae on pro- lateral trochanter and femur of leg I and on the retrolateral coxa and trochanter of palp (in some species these setae are finely plumose); no Targe’ plumose (lanciform or spatulate) se- tae such as those on the prolaterobasal femur of leg I in Euathlus or those on the prolateral coxa of leg I in Grammostola; (3) type I urti- cating hair only; (4) corresponding segments of all legs approximately the same width in fe- males; femur III in males of some species lat- erally swollen; (5) scopula of tarsus IV usually 148 THE JOURNAL OF ARACHNOLOGY entire, if divided then only partially and nar- rowly by line of setae; (6) setae of prolateral coxa I hairlike and not basally swollen (known only in small species), spiniform and basally swollen, or distinctly thomlike (apically fili- form), with all forms at least distally plumose; (7) metatarsus I flexing against lower process of tibial spur, with either apex of spur contact- ing ventral surface of metatarsus or outer edge of spur in the apical half contacting the prola- teral metatarsus; (8) lower (outer) process of tibial spur curving prolaterodistally and wid- ening apically, usually equipped with at least one apical or preapical megaspine, and upper (inner) shorter process less stout basally, rela- tively uniform in diameter throughout its length, and equipped on its inner surface with at least one (several not uncommon) stout, ba- sally articulated megaspine. Additional diagnostic characters. — Based on preliminary data, (1) ventral retro- marginal setae of maxillae and ventral mar- ginal setae of coxae (other than distal mar- gins) similar in form (and usually size) to prolateral setae of coxa I, or, similar to other ventral setae of coxae (more common); (2) extent of metatarsus IV scopula usually from distal 20-85%, rarely less than 20% or with scattered scopula hairs (A. paloma and undescribed species from southeast Arizo- na); (3) anterior sternal margin smoothly procurved or rarely with a broad but slight medial projection; (4) labiosternal suture re- cessed; labium rising steeply from suture; (5) male embolus tapering with inward and ventral curve, with strong ventrally directed bend near apex (in all species with slender emboli), embolus either very slender with three apical keels (medial keel most promi- nent), or, relatively wide with four promi- nent keels of which the distomedial and proximomedial keels are either convergent or closely parallel just basad of apex, with proximomedial or convergent keel serrate and/or extending the full or nearly full length of embolus in some species; (6) tran- sition of valley between processes of tibial spur considerably offset (protruding) from longitudinal plane of tibia; (7) tibiae I and II of females with at least one ventral spine (other than apical), rarely none and then only in species where male embolus is rel- atively broad apically; usually also with at least one prolateral spine, rarely absent on tibia II; (8) paired spermathecae separated and with capitate bulbs, variable in shape. Genera removed from synonymy.— Two genera placed in the synonymy of Aphono- pelma, Clavopelma and Pterinopelma, differ from the remaining Aphonopelma in char- acters here considered generic in value (above). Clavopelma Chamberlin (monotyp- ic genus) is here removed from the synon- ymy of Aphonopelma because of the follow- ing differences: both types I and III urticating hairs, lanciform setae (similar to those in Euathlus (Brachypelma) but rela- tively smaller) on the prolateral trochanter and femur of leg I and on retrolateral tro- chanter of the palp, and a relatively straight (slender) embolus without a sharp ventrally directed bend near apex. Pterinopelma Po- cock, as defined, is similar to Euathlus in the form of scopular setae on the posterior face of the palpal trochanter and the anterior face of the trochanter of leg I but dissimilar to both Euathlus and Aphonopelma in lack- ing a similar scopula of hairs on the anterior face of femur 1. The presence of both types I and III urticating hairs, characteristic of Euathlus, is also characteristic of Pterino- pelma. The anterior lateral eyes of Pterino- pelma are proportionately larger relative to AME than in other genera in the synonymy of Aphonopelma. Pterinopelma is here not considered a synonym of Aphonopelma but its final disposition is reserved until all Pter- inopelma species have been examined. PARTIAL KEY TO MALES OF APHONOPELMA Phylogenetic relationships are not implied by the couplets. Also, the asterisk (*) indicates that key is for types only, and does not account for variation. 1. Setae on prolaeral coxa I hairlike, not basally swollen (Figs. 7, 8); small species ......... 2 Setae on prolateral coxa spiniform, basally swollen (Figs. 9, 10). .................... . 5 PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 149 Figures 1-4. — Sternal and coxal setae, males (arrows indicate setae). 1, 2, Aphonopelma Joshua new species; 1, Sternum, medial; 2, Coxa I, ventral, proximo-medial; 3, 4, Aphonopelma mojave new species; 3, Sternum, medial; 4, Coxa I, ventral, proximo-medial. 2(1). Tarsus IV scopula as least partially divided by setae (Fig. 5), ..... 3 Tarsus IV scopula complete, not divided by setae (Fig. 6) 4 3(2). Medial sternal setae hairlike or attenuate (refer to Fig. 3); Arizona ................ .paloma Medial sternal setae short, stout, basally thickened, and conicform (Fig. 1) Joshua new species 4(2). Length ratios tibia I/metatarsus I, 0.92-1.07, femur I/tibia I, 1.1 8-1 -.28 (Table 1) ....... . mojave new species Both ratios = 1.10 (Table 1); California, Manhattan Beach (type locality doubtful) . . . *radmum 5(1). Extent of metatarsus IV scopula less than distal 40% .............................. 6 Metatarsus IV scopula at least distal 55% 7 6(5). Carapace length >13 mm; California, Baja; in life carapace and legs medium brown to black ......................................................... *steindachneri Carapace length <10 mm; Utah, New Mexico (stimulatum) = *marxi 7(5). Scopula metatarsus IV, distal 55-65% 8 Scopula metatarsus IV > distal 70% .......................................... 10 8(7). Palpal bulb retrolateral bend uniform (as in Figs. 14, 29) 9 Bulb with retrolateral bend angular (as in Fig. 18); Grand Canyon, Phantom Ranch . . *phasmus 9(8). Length carapace <10 mm; length ratio leg IV/carapace approximately 3.6; Arizona .... *zionis Length carapace >15 mm; length leg IV/carapace approximately 3.0; Baja *helluo 10(7). Length ratio leg IV/carapace = 3.09-3.60; California iodium Ratio leg IV/carapace <3.0; Arixona, Mexico ............................... *rusticum 150 THE JOURNAL OF ARACHNOLOGY PARTIAL KEY TO FEMALES OF APHONOPELMA Phylogenetic relationships are not implied by the couplets. Also, the asterisk (*) indicates that key is for types only, and does not account for variation. 1. Setae on prolaeral coxa I hairlike, not basally swollen (Figs. 7, 8); small species ......... 2 Setae on prolateral coxa spiniform, basally swollen (Figs. 9, 10) 4 2(1). Tarsus IV scopula partially divided by setae (Fig. 5) .............................. . 3 Tarsus IV scopula complete, not divided by setae (Fig. 6) mojave new species 3(2). Metatarsus IV scopula >30 distal percent and almost always <50% ...... Joshua new species General reduction of metatarsal scopulation, metatasus scopula <20%; Arizona. .paloma 4(1). Metatarsus IV scopula < 35%; carapace and legs dark in color (Table 2) *steindachneri Metatarsus IV scopula > 70%; carapace and patellae and tibiae I, II, pale buff (Table 2) . . . iodium Aphonopelma Joshua new species Figs, 1, 2, 5, 7, 12=15, 22, 23, 51; Map 1. Types. — Holotype male from San Bernar- dino County, allotype female from Riverside County, California, both from the Covington Flats area of Joshua Tree National Monument. Holotype collected at 10:09 PM, 6 September 1992, 2.3 mi. below the Covington Flats en- trance to JTNM, elevation 3660 ft. Allotype excavated from a mounded burrow 21 Octo- ber 1989, 5.6 mi. into the Monument in the Upper Covington Flat area, elevation 5140 ft. Paratype males (12): 24 July 1989 (1), 3 Au- gust 1989 (1), 10 August 1989 (2); 27-28 July 1990 (2); 27-28 July 1992 (5), 12 August 1992 (1). Paratype females (2): 3 May 1989 (1); 30 July 1992 (1). All paratypes collected by author in the Covington Flats area or near the JTNM entrance to this area. Types depos- ited in AMNH. Etymology. — The specific name is a noun in apposition taken from the type locality, Joshua Tree National Monument. Diagnosis. — A. Joshua new species can be distinguished from all other species by the fol- lowing combination of characters: small size, hair-like form of setae on prolateral coxa I (Fig. 7), partial division of tarsus IV scopula by setae (Fig. 5), and extent of metatarsus IV scopula. Males (Table 1) are most easily rec- ognized by their unique conicform setae of the sternum (Fig. 1), maxillae (similar to sternal setae), and coxae (Fig. 2) and the laterally swollen third femur (Fig. 13). They are sep- arated from males of the most similar species, A. mojave new species, A. radinum and A. paloma, by the following: from A. mojave and A. radinum by partial division of tarsus IV scopula (Fig. 5) and form of the palpal bulb (Figs. 14, 15) and from A. paloma by more extensive scopula of metatarsus IV (also III), respectively. Females (Table 2) are distin- guished from those of all other species except A. paloma by partial division of tarsus IV scopula and from A. paloma by the more ex- tensive scopula of metatarsus IV (all metatar- si). Description. — Male: Holotype. Overall length, 23.80; carapace, length, 8.80 width, 7.65; chelicerae, length, 2.80, width, 3.70. Chel- iceral macroteeth, 8, denticles, 14 right, 15 left; sternum, length, 3.80, width, 3.60. Labial cus- pules, 73; maxillary cuspules, 89 right, 98 left. Color of entire tarantula appears black; carapace pubescence black with a silver or gray-black sheen, appressed and moderately dense; chelic- erae black with a silver sheen; abdominal pu- bescence gray-black, somewhat lighter than car- apace. Abdominal anterodorsal setae black and relatively stout; lateral setae black, slender and shorter than dorsal setae; ventral setae black and finer than lateral setae; circular patch of dark black, type I (Cooke et al. 1972) urticating hairs (as in Fig. 11) covering posterodorsal half of abdomen, clearly visible; longest setae hairlike, pale orange-buff, basally dark, interspersed mostly within, just outside and caudally below uiticating patch. Legs with black pubescence (legs appear gray-black), femora slightly darker; longest leg setae black with distal half pale or- ange-buff, other leg setae black. Ventral aspect black except red-orange labium and anterior pal- pal coxae and orange scopulae of palpal coxae. Cephalic region of cai'apace rising gradually PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 151 Table 2. — Females of Aphonopelma: Taxonomic characters and quantitative character values which separate the Mojave Desert Aphonopelma and A. paloma. ° — species sharing the hairlike form of setae on prolateral coxa I, distinguished by superscript only from A. Joshua (') and A. mojave (^). * — include in the synonymy of A. iodium; ' — no overlap with A. Joshua-, ^ — no overlap with A, mojave-, ^ — no overlap with A. iodium', gen dedu all — general reduction of all metatarsal scopulae; other abbreviations defined in Methods section of text. Mean and standard deviation shown in parentheses, i.e., (7.76, 1.34). Carapace measurements are in millimeters. °Joshua [n = 10] °moJave [n = 30] iodium [n = 14] ° paloma [n = 11] *angusi [allotype] LC 6.00-9.70 (7.76, 1.34) 6.00-10.25 (8.03, 1.03) 2' 10.80-22.05 (15.02, 2.65) 4.10-6.10 (5.05, 0.57) ^9.10 LFI/LTI 1.26-1.32 (1.29, 0.02) 1.27-1.36 (1.31, 0.02) 1.32-1.38 (1.36, 0.02) 1.29-1.39 (1.35, 0.03) 1.32 LFI/LTII 1.44-1.50 (1.48, 0.02) 1.48-1.60 (1.54, 0.03) 1.50-1.57 (1.53, 0.02) 2' 1.63-1.74 (1.68, 0.04) '1.56 LFI/LMI 1.41-1.55 (1.49, 0.05) 1.41-1.55 (1.48, 0.04) 1.29-1.47 (1.40, 0.04) 2'1.56-1.71 (1.64, 0.04) -n.52 LFI/LMII 1.40-1.57 (1.50, 0.06) 1.49-1.63 (1.55, 0.04) 1.33-1.52 (1.46, 0.05) 2'1.70-1.82 (1.76, 0.05) n.56 LMIII/LMI 1.09-1.19 (1.15, 0.03) 0.99-1.10 (1.05, 0.03) '1.01-1.07 (1.04, 0.02) '1.00-1.07 (1.02, 0.30) ' '1.07 LALLC 2.26-2.43 (2.37, 0.06) 2.21-2.74 (2.39, 0.10) 2.23-2.51 (2.42, 0.09) '1.94-2.23 (2.13, 0.09) 2.39 LAIV/LC 2.59-2.84 (2.75, 0.07) 2.36-3.06 (2.63, 0.19) 2.41-2.70 (2.64, 0.11) 2' 1.94-2.23 (2.13, 0.10) 2.68 LAI/LAIV 20.85-0.88 (0.86, 0.01) '0.89-0.94 (0.91, 0.01) '0.89-0.93 (0.92, 0.01) 0.88-0.94 (0.90, 0.02) '0.89 LAI/LAIII 1.08-1.15 (1.12, 0.02) 1.14-1.23 (1.18, 0.03) 1.11-1.15 (1.14, 0.01) '1.22-1.27 (1.25, 0.02) ^'1.16 Scopula 30-65 25-50 2 '75-85 (lat) 2'0-17 2' >70 (worn) MIV (%) (40) (35) 45-60 (med) gen redu all 45 (med) Division Sc TalV (%) 2par div 40-70 'undivided 'undivided 2divided 40-67 'undivided Prolateral hairlike hairlike 2'spiniform hairlike 2'spiniform Cxi setae bas uniform bas uniform bas swollen bas uniform bas swollen Carapace color black black 2 'paper-bag brown black 2 'paper-bag brown Legs I, II, Palp (Ti, Pt color) black black 2 'paper-bag brown black 2 'paper-bag brown from thoracic groove, slightly less than twice the height of thoracic region. Ocular turret rel- atively high, compact, and steep resulting in a more lateral rather than dorsolateral inchnation of lateral eyes, ocular area width 0.28 X maxi- mum width of cephalic region. AME circular, AME-AME, 0.6 X AME diameter, AME- ALE, 0.15, 0.20X (left, right, respectively) AME di- ameter, AME-PME, 0.05, O.lOX AME diame- ter; ALE, PLE roughly ovoid (somewhat flat- tened ventrally), ALE length, l.OOX AME diameter, ALE-PLE, 0.20, 0.30X AME diame- ter, PLE, 0.90, 0.95 X AME diameter, PLE-PME contiguous; PME ovoid, 0.70, 0.80X AME di- ameter. Thoracic groove a transverse pit with anterior border straight. Sternum widest be- tween bases of coxae n & HI; unique medial sternal setae short, basally swollen, and sharply constricted toward apex, some with hairlike api- cal portion but in most appear to be broken off; 152 THE JOURNAL OF ARACHNOLOGY Map 1. — Distribution of Aphonopelma Joshua new species (A) and Aphonopelma mojave (o). The boundries of the Mojave Desert (as perceived by the authjor) are indicated by the outer-most dotten lines; the area delimited by the inner-most dotted line indicates the geographic barrier that separates eastern and western populations of A. mojave. marginal setae stout, slightly swollen basally, longer than medial setae; intermediate setae similar to marginal setae but less stout. Medial coxal setae (coxae I-IV) stout, thickened basally, similar* to medial sternal setae although more elongate; distal and mai*ginal setae similar to marginal sternal setae, proximal setae similar to intemiediate and medial sternal setae. Distal maxillary setae similar to mai'ginal sternal setae; basomarginal, retromarginal, and medial setae similar to medial sternal setae but slightly less stout and usually apically filiform; most anterior setae relatively fine, hair-like. Femur HI laterally swollen, at widest point 2.35, femur I, 1.70 (widest point other than at base just preapical of articulation with patella), WFTIIAVFI = 1.36. Prolateral face of coxa I with pad of fine hair- like, distally plumose setae both above and be- low suture. Tibia I arcuate, somewhat less than average condition. Leg and palp segment lengths in Table 3. Extent of scopulae (XlOO = percent): metatarsi I & n complete; metatarsus III, prolaterodistal 0.75 (0.95 if scattered indi- vidual scopula hairs are considered), retrolater- odistal, left, 0.60, right, 0.55; metatarsus IV, 0.40 mediodistal. Metatarsus IV scopula com- pletely divided by setae; tarsus IV scopula di- vided by setae proximal, left, 0.50, right, 0.60. Spination: metatarsus I, Iv(am), tibia I, L2d(lp0.30 lp0.70) Rld(lp0.30) L5v(lr0.10 ler lr0.50 lpO.45 lpO.55) R3v(ler lp0.50 lrO.55), femur I, ld(p0.85); metatarsus II, L4v(lap lam lar lr0.30) R2v(lam li-0.35), tibia II, 2d(10.33 lpO.67) L4v(lap lar lr0.50 lrO.55) R5v(lap lar lp0.50 lr0.15 lrO.55), patella 11, Lld(p0.50), femur II, ld(p0.85); metatarsus III, 4d(lep ler lpO.35 lr0.40) L6v(lap lam 2ar lpO.45 11*0.20) R7v(lap lam lar lp0.50 lr0.20 lrO.45), tibia III, L4d(lp0.25 lpO.65 lr0.15 lr0.90) R4d(lp0.60 lr0.20 lr0.60 lr0.90) L5v(2ap lar lp0.50 lr0.50) R4v(lap lar lp0.50 li*0.45), femur HI, Rld(lr0.80); metatarsus IV, L4d(lep ler lp0.40 lr0.40) R3d(lep ler lrO.45) L14v(lap lam 2ar 3p0. 10-0.50 7r0. 15-0.65) R14v(lap lam lar 2p0.30-0.50 8r0. 10-0.80), Figures 5, 6. — Tarsus IV scopulae. 5, Aphonopelma Joshua new species, division of scopulae by setae (setae indicated by arrows); 6, Aphonopelma mojave new species, scopula undivided. PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 153 Table 3. — Aphonopelma Joshua new species, ho- lotype male: leg and pedipalp segment lengths. I II III IV Palp Femur 9.20 8.75 8.40 9.80 5.20 Patella 4.25 3.90 3.60 3.80 2.80 Tibia 8.00 7.50 6.90 8.50 4.65 Metatarsus 8.20 8.30 9.60 11.75 Tarsus 5.30 5.30 5.40 6.00 1.90 Total length 34.95 33.75 33.90 39.85 14.55 tibia TV, Lld(lr0.40) R3d(lp0.65 lrO.45 lrO.75) L5v(lap lar lpO.50 IrO.lO lr0.50) R4v(lap lar lpO.65 lrO.55); palpal tibia, 3d(lp0.35 lp0.60 lpO.85) L4v(lp0.55 lp0.80 lpO.85 lrO.50) R3v(lp0.55 lpO.80 lrO.45), palpal patella, Lld(p0.50), palpal femur, ld(p0.85). Female: Allotype. Overall length, 28.0; car- apace, length, 9.70, width, 8.70; sternum, length, 4.35, width, 4.00; chelicerae, length, 3.70, width, 5.30. Cheliceral macroteeth, 7 right, 8 left; denticles, 9 right, 12 left. Labial cuspules, 50; maxillary cuspules, 72 right, 79 left. General color dark gray-black or black; carapace with bronze or gray-green sheen, pubescence ap- pressed, medium density; chelicerae black with silver sheen; abdominal pubescence mouse gray or gray-black, oblong patch of urticating hairs black, clearly visible; legs gray, darker than ab- domen, transitional between abdomen and car- apace colors; ventral aspect dark gray-black (ab- domen slightly less black) except orange color of labium, palpal coxae and scopulae of palpal coxae. Abdominal anterodorsal setae black, rel- atively long, with distal portion pale buff, less stout than homologous setae in male; long, ba- sally dark pale orange-buff setae generally lon- ger and slightly more slender than anterodorsal setae, interspersed mostly within and just out- side of urticating patch; anterolateral, lower pos- terolateral, and ventral setae generally shorter Figures 7-10. — Prolateral setae of coxa I (arrows indicate setal bases). 7, Hairlike setae of Aphonopelma Joshua new species; 8, Hairlike setae of Aphonopelma mojave new species; 9, Spiniform, basally thick- ened setae, Aphonopelma iodium. Red Mountain, California; 10, Spiniform, basally thickened setae, Aphonopelma iodium, Beaver Dam Mountains, Utah. 154 THE JOURNAL OF ARACHNOLOGY and more slender than dorsal setae, similar in color to anterodorsal setae; dark ventral setae often very fine. Oblong patch of black urticating hairs covering posterodorsal 60% of abdomen. Legs with gray-black pubescence, shortest setae mostly black, longer setae with proximal half black, distal half pale orange-buff. Cephalic re- gion of carapace rising from thoracic region at a steeper slope than in male, almost three times height of lowest area of thoracic region. Ocular tun'et intermediate in height, width 0.24 X max- imum cephalic region width; lateral eyes with relatively normal dorsolateral orientation. AME circular, AME- AME, 0.9 X AME diameter, AME- ALE, 0.4 X AME diameter, AME-PME, 0. 15X AME diameter; ALE and RLE roughly ovoid, somewhat flattened ventrally; ALE length, l.OX AME diameter, ALE-PLE, 0.5 X AME diameter; PLE length, 0.6, 0.7 X AME di- ameter, PLE-PME almost contiguous; PME sub- circular, 0.5 X AME diameter. Thoracic gi'oove transverse, slightly procurved. Sternum widest between bases of coxae II & III. Contraiy to condition in male all sternal setae relatively fine and attenuate, medial setae generally longer but relatively slender compared to stout, spiniform marginal setae. Medial setae of coxae HI & IV and most basal, promarginal, and reti*omarginal setae of all leg coxae intermediate between me- dial and marginal sternal setae in basal diameter; medial setae of coxae I & II similai* to medial sternal setae. Basomarginal, retromarginal, and distal setae of palpal coxae inteiTnediate in basal diameter, setae becoming more slender toward promargin. Femur HI not swollen as in male. All leg segments shorter than cai*apace length; femur IV and metatai*sus IV longer than femur 1. Setation on prolateral face of coxa I as in holotype. Leg and palp segment lengths are in Table 4. Extent of scopulae (XlOO = %): meta- tarsi I & II scopulae as in holotype; metatarsus III, prolaterodistal 0.60, retrolaterodistal 0.50; metatarsus IV, prolaterodistal 0.35, retrolater- odistal 0.25. Metatarsus IV scopula divided by setae; tarsus FV scopula divided by setae prox- imal 55 percent. Spination: metatarsus I, Iv(am), tibia I, L2d(lp0.25 lp0.60) (R2d(lp0.20 lp0.60) L5v(2ap lar lp0.15 lpO.45) R5v(2ap lar lp0.15 lp0.40), patella I, ld(p0.50) Rlv(m0.75), femur I, ld(p0.80); metatarsus II, Rld(p0.35) 3v(lap lam liH.35), tibia n, L2d(lp0.25 lp0.60) L5v(2ap lar lp0.15 lpO.45) R3d(lp0.20 lp0.60 lpO.85) R6v(2ap lai- lpO.35 lr0.15 lr0.40), patella H, ld(p0.50) Table 4. — Aphonopelma Joshua new species, al- lotype female: leg and pedipalp segment lengths. I II III IV Palp Femur 7.70 7.00 6.50 8.20 5.60 Patella 4.00 3.80 3.50 3.80 3.10 Tibia 5.90 5.20 4.70 6.40 4.20 Metatarsus 5.40 5.30 6.00 8.05 Tarsus 4.00 4.00 4.20 4.70 4.00 Total length 27.00 25.30 24.90 31.15 16.90 Iv(m0.85), femur H, ld(p0.75); metatarsus IH, L5d(lp0.15 lpO.35 lepO.85 lr0.40 ler0.80) L6v(lap lam lar lpO.25 lp0.40 lr0.40) R6d(lp0.15 lpO.45 lepO.85 lrO.45 2er0.85) R5v(lap lam 2ar lr0.40), tibia HI, L6d(lp0.20 lpO.55 lpO.85 lr0.15 lrO.55 lr0.80) L6v(2ap lar lp0.40 lr0.15 lr0.50) R5d(lp0.25 lpO.55 lrO.25 lrO.55 lrO.85) R10v(2ap lar lp0.40 lp0.70 lr0.15 lrO.25 lr0.40 lr0.60 lrO.65), fe- mur III, L2d(lr0.70 lrO.85) R2d(lp0.80 lr0.80); metatarsus IV, L4d(lp0.40 lep0.90 lrO.45 ler0.90) Lllv(lap lam 2ar lp0.30 lpO.55 5r0. 15-0.70) R4d(lp0.45 lepO.85 lr0.40 lr0.90) R14v(lap lam 2ar lp0.15 lp0.30 lp0.50 7r0.15-0.85), tibia IV, L4d)lp0.65 lr0.15 lrO.65 lrO.85) L7v(2ap lar lpO.45 lr0.15 lrO.36 lr0.60) R4d(lp0.65 lr0.15 \xQ.65 110.85) R6v(2ap lar lpO.45 lr0.15 lr0.40), fe- mur IV, Lld(r0.75); palpal tibia, 2d(lp0.55 lp0.90) L8v(3ap lar lpO.25 lp0.60 lrO.45 lerO.85) R9v(2ap lar lpO.25 lp0.60 lr0.20 lrO.45 lr0.60 ler0.90), palpal patella, ld(p0.25), palpal femur, ld(p0.85). Variation. — Males: Total length 19.00- 26.75. Cheliceral macroteeth, 7-9, 8 most common (60%), 9 least common (6%), den- ticles, 5-17. Labial cuspules, 33-78, x = 54; maxillary cuspules, 69-117 (each side), x = 88. Coloration of new males tends to fade over time to dark brown-gray, carapace often with a bronze sheen. In some specimens the long pale orange-buff setae, normally inter- spersed within or just outside urticating patch, are sparsely interspersed on the venter, lateral surfaces, and slightly more anteriad of urti- cating patch. Patch of type I urticating hair (as in Fig. 11) covering distal 40-60% of abdo- men. Femur I and metatarsus III almost al- ways longer than, raiely equal to or shorter than, carapace; tibia I usually shorter than, rarely equal to or longer than, metatarsus I; metatarsus I slightly to moderately arcuate PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 155 Table 5. — Aphonopelma Joshua new species, males (25 including holotype): range of leg and pedipalp segment lengths. I II III IV Palp Femur 7.85-9.85 7.40-9.40 7.20-9.00 8.35-10.50 4.30-5.60 Patella 3.60-4.50 3.40-4.20 3.00-3.80 3.20-4.00 2.30-2.90 Tibia 6.75-8.40 6.10-7.80 5.90-7.20 7.30-8.90 4.00-4.95 Metatarsus 6.85-9.00 6.80-9.20 8.10-10.55 10.20-12.70 Tarsus 4.40-5.60 4.40-5.60 4.50-5.80 5.10-6.50 1.50-2.00 proximal portion of segment (in comparison to A. mojave); femur III laterally swollen (Figs. 12, 13), variable in degree. Ranges of leg and pedipalp segment lengths are in Table 5. Basal division of palpal bulb variable in shape; bulb itself and embolus (Fig. 14) rel- atively constant in shape; proximal prolateral protuberance of middle division slight (Fig. 15) or absent. Extent of scopulae (XlOO = %): metatarsus II complete prolaterally but of- ten very sparse or short of base retrolaterally and medially, metatarsus III prolaterodistal 0.50-0.85, X = 0.63 (often scattered hairs past 0.75), slight retrolateral reduction metatarsus IV 0.25-0.50, X ^ 0.36 (medial). Ventroapical metatarsal spination: I, 1-2, II, 2-3, III, 3-5 (usually 4), IV, 4-6 (usually 4). Females: Total length 15.70-28.00. Chelic- eral macroteeth, 7-9, 8 most common (70%), 9 least common (20%), denticles, 5-25 (each side). Labial cuspules, 50-97, x = 69; max- illary cuspules, 67-154 (each maxilla), x = 93. Coloration of females tends to fade over time, carapace and chelicerae to bronze, gray- green, or brown-gray (chelicerae somewhat darker), abdomen to mouse gray or faded car- apace color (usually lighter), and legs to brown-gray (intermediate in color between carapace and abdomen) with femora and tarsi darker. In some females, the long, pale or- ange-buff abdominal setae are sparsely inter- spersed more anteriad of the urticating patch, on the venter, and on the lateral abdominal surfaces. Spermathecae variable in shape and relative distance between bulbs (Figs. 22, 23). Although spermathecal characters have been found useful in distinguishing between other closely related mygalomorph species they ap- pear to be of no diagnostic value in separating females of A. Joshua from those of A. mojave (Figs. 24-28). Range of leg and pedipalp seg- ment lengths in Table 6. Extent of scopulae (XlOO == %): metatarsus III, prolaterodistal 0.55-0.85, X = 0.65 (only one specimen great- er than 0.75), slight retrolateral reduction, metatarsus IV, distal 0.30-0.65 (usually me- dial), X = 0.40, lateral extent slightly reduced, prolateral extent usually greater than retrola- teral. Ventroapical metatarsal spination: I, 1- 2, II, 2-3, III, 4-5 (rarely 5), IV, 4. Figures 11-13. — 11, Type I urticating hair, {Aphonopelma mojave new species, male, east Mojave); 12, 13, Femora of Aphonopelma Joshua new species showing comparative widths; 12, Femur I (normal); 13, Femur III (swollen). 156 THE JOURNAL OF ARACHNOLOGY Figures 14-21. — Palpal bulbs of Aphonopelma Joshua new species and Aphonopelma mojave new species, right; even, ventral (short arrows show retrolateral bend into apical portion); odd, dorsal (long arrows show position and degree of protrusion of proximal prolateral protuberance). 14, 15, A. Joshua; 16, 17, A. mojave, east Mojave, near Kelso, California; 18, 19, A. mojave, east Mojave, BDM, Utah; 20, 21, A. mojave, west Mojave, Red Mountain, California. Distribution. — A. Joshua has a relatively limited distribution, primarily, in Joshua Tree National Monument between the northern flanks of the Eagle, Cottonwood and Little San Bernardino Mountains and the southern flanks of the Pinto Mountains (excluding the eastern Pinto Mountains and Pinto Basin), Queen Mountain, and Wonderland of Rocks. Outside of the monument, the species occurs south of the juncture of the Eagle and Cotton- wood Mountains in a very limited area above 550 m in elevation, west of the northwestern boundary of the Monument in upper Morongo Valley, and north of Yucca Valley in two foot- hill desert valleys (San Bernardino Moun- tains), one serviced by Pipes Canyon Rd. (roughly parallels Pipes Wash), the other by New Dixie Mine Road, approximately 10 km north of Pipes Canyon Road, south of the Big- horn Mountains. This latter area appears to be the northern-most limit of the species. A. Josh- ua would be considered rare north of a diag- onal connecting Queen Mountain and New Dixie Mine Rd. The distribution of A. Joshua is shown on Map 1. Material examined.™ Type specimens and the following: CALIFORNIA: Riverside County: JTNM, Fried Liver Wash, 27 August 1965 (E.L.'s. & S.L.J.), 19. Squaw Tank, 3500 ft. elev., IS, 9 September 1966 (E.L. Sleeper & S.L. Jenkins). Pleasant Valley, ;lma, 23 September 1967 (E.L. Sleeper & S.L. Jenkins). Lost Horse Valley, 1.1 mi. S of Quail Springs Rd. on Keys View Rd., 1 mi. W of K. V. Rd., 4383 ft. elev., definitive molt, 7 July 1989; IS, 3 May 1989. 4400 ft. elev., 19, Cottonwood Springs, near visitor’s center, 3100 ft. elev., 3S, 23 August 1989; Cottonwood Springs Rd, Smoke Tree Wash, 4.4 mi. N of visitor’s center, 2710 ft. elev., 1(3, 31 August 1989. San Bernardino County: JTNM, west of Wonderland of Rocks, 4.5 mi. SE of monument entrance on Quail Springs Rd., 3950 ft. elev., definitive molt,7 July 1989; 1(3, 28 March 1989; Pipes Canyon Rd. (Pipes Wash), 4.5 Table 6. — Aphonopelma Joshua new species, females (10 including allotype): range of leg and pedipalp segment lengths. I II III IV Palp Femur 4.90-7.70 4.40-7.00 4.00-6.50 5.15-8.20 3.60-5.60 Patella 2.60-4.00 2.40-3.80 2.20-3.50 2.40-3.80 2.00-3.10 Tibia 3.90-5.90 3.30-5.20 2.90-4.70 4.20-6.40 2.70-4.20 Metatarsus 3.20-5.40 3.15-5.30 3.60-6.00 5.10-8.05 Tarsus 2.50-4.00 2.50-4.00 2.65-4.20 3.10-4.70 2.60-4.00 PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 157 mi. W of Hwy 247, 4270 ft. elev., definitive molt, 24 June 1991; Id, 4 November 1989; 4300 ft. elev., Id, 1 August 1992; 2.5 mi. W of Hwy 247, 4000 ft. elev., definitive molt, July 1992; Id, 18 April 1990; 8.3 mi. W of Hwy 247 toward Burns Canyon, 4350 ft. elev., 3 2,2 Nov. 1991; 5.9 mi. W of Hwy 247 toward Bums Canyon Rd., 4240 ft. elev., 1 2 , 13 Sept. 1992; New Dixie Mine Rd., 10.5 mi. N Yucca Valley on Hwy 247, 6.3 mi. W, 5000 ft. elev., 12,11 Sept. 1992; Morongo Valley, 5.1 mi. NE of Post Office off Hwy 62, 2890 ft. elev., ;1 2, 6 Sept. 1993. Specimens collected by the author deposited in AMNH. Aphonopelma mojave new species Figs. 3, 4, 6, 8, 11, 16-21, 24-28; Map 1 Types. — ^Holotype male, allotype female, 9 paratype males, and 5 paratype females from San Bernardino County and Kem County Cal- ifornia, south of Red Mtn., 20 mi. N of Kra- mer Jet. on Hwy 395, 1-2 mi. W of Hwy 395. Holotype collected 28 October 1989, 12:45 PM, 3450 ft. elev. Allotype lured out of bur- row after dark 20 October 1991, 3250 ft. elev. Paratype males: 1974, 10 October 1989 (2) (W. Icenogle); 28 October 1991 (1); 14 Oc- tober (1), 26 October (5), 2990-3450 ft. elev. Paratype females: 13 October 1991 (1), 20 October 1991 (1); 1992, 25 January 1992 (1), 4 October 1992 (1), 13 October 1992 (1), 3220-3280 ft. elev. All types except those specified collected by author. Types deposited in AMNH. Etymology. — The specific name is a noun in apposition taken from the name of the des- ert within which the species appears to be al- most totally contained. Diagnosis. — A. mojave new species is dis- tinguished from other species by the following combination of characters: carapace color, hair-like setae of prolateral coxa I (Fig. 8), undivided tarsus IV scopula (Fig. 6), limited extent of metatarsus IV scopula, and propor- tional lengths of tibiae and metatarsi I and II and leg III. In only three other species, A. Joshua, A. paloma, and A. radinum, are the prolateral setae of coxa I hair-like. A. mojave is easily distinguished from the two former species by entire scopula of tarsus IV and again from A. Joshua by the spiniform setae of the sternum (Fig. 3) and ventral coxae (Fig. 4) in males and from A. radinum (female un- known) by proportionately shorter tibiae I and II and longer leg III. In one other similar spe- cies, A. marxi {= A. simulatum), the prolateral setae of coxa I are slightly swollen basally; A. mojave is distinguished from this species (fe- males unknown) by proportionately much lon- ger metatarsi I and II and legs III and IV. A. mojave is easily distinguished from two other small species, A. phasmus and A. zionis, by characters in Table 1. Description. — Male: Holotype. Total length, 19.70; carapace, length, 8.50, width, 7.50; ster- num, width, 3.65, length, 3.70; chelicerae, width, 3.80, length, 2.50. Cheliceral macro- teeth, right 9, left 8; each side with 10 denti- cles. Labial cuspules, 33; maxillary cuspules, 90 right, 95 left. General color black with a faint bluish sheen; carapace with black pubes- cence, not appressed, moderately dense in ce- phalic region, increasing in density toward pos- terolateral and caudal margins. Abdomen clothed with blue-black pubescence; long, ba- sally dark, orange-buff (or orange-tan) setae in- terspersed over entire dorsal, posterolateral, and caudal surfaces; slightly shorter versions of these setae on anterolateral and ventral sur- faces, least dense on venter; circular patch of black type I urticating hairs (Fig. 11) covering posterodorsal 45 percent of abdomen, not clearly visible because of pubescence colora- tion and extensive interspersion of long orange- buff setae. Leg pubescence black, longer setae similar in color to abdominal setae, shorter se- tae dark with pale orangish-buff apices. Ce- phalic region rising gradually from thoracic re- gion, slightly more than one and a half times higher. Ocular turret width slightly greater than 20 percent of maximum cephalic width, inter- mediate in height. AME circular, approximate- ly 2.5, AME- AME, 0.6 X AME diameter, AME- ALE, 0.15X AME diameter, AME-PME, O.IX AME diameter; ALE roughly ovoid, somewhat flattened ventrally, 0,9 X AME di- ameter, ALE-PLE, 0.4 X AME diameter, ALE- PME, 0.5 X AME diameter; PLE, subcircular, 0.55 X AME diameter, PLE-PME contiguous; PME irregular to elongate ovoid, slightly lon- ger and narrower than PLE. Thoracic groove a transverse pit with anterior edge procurved. Medial sternal setae slender, attenuate; margin- al setae basally stout, more spiniform; setae be- tween medial and marginal setae intermediate in basal diameter. Promarginal and retromar- ginal setae of coxae I-IV similar in basal di- ameter to marginal sternal setae; medial setae similar to intermediate sternal setae, Baso- and retromarginal setae of palpal coxa similai" to 158 THE JOURNAL OF ARACHNOLOGY Table 7. — Aphonopelma mojave new species, holotype male: leg and pedipalp segment lengths. I II III IV Palp Femur 8.20 7.70 7.10 8.35 4.90 Patella 3.85 3.60 3.15 3.40 2.60 Tibia 6.60 5.95 5.30 6.85 4.50 Metatarsus 6.60 6.55 7.10 9.00 Tarsus 4.50 4.50 4.50 5.00 2.00 Total length 29.75 28.30 27.15 32.60 14.00 intermediate sternal setae; distal setae similar to marginal sternal setae, setae becoming finer toward promargin of coxa. Femur III only slightly wider than femur I, WFIIIAVFI = 1 .09. Metatarsus I basally arcuate, bend moderate. Prolateral face of coxa I with fine, hairlike, dis- tally plumose setae both above and below su- ture. Leg and palp segment lengths in Table 7. Extent of scopulae (XlOO = %): metatarsi I and II to base (retrolateral scopula of metatar- sus II, distal 0.85), metatarsus III, prolaterodis- tal 0.70, 0.75 (right and left, respectively), ret- rolaterodistal 0.55, 0.60, metatarsus IV, distal 0.40 medial, 0.25 lateral. Tarsus IV scopula en- tire, not divided by setae; metatarsus IV scop- ula divided proximal, right, 25, left, 40 percent. Spination: metatarsus I, Iv(am), tibia I, 2d(lp0.30 lp0.70) R4v(lbr lr0.20 11*0.50 ler) L4v(lbr lr0.50 2er), femur I, ld(ep>0.80); metatarsus II, R3v(lr0.30 lap lam) L2v(lap lam), tibia II, 2d(lp0.25 lpO.65) R5v(lbr lrO.35 2ap lar) L5v(lbr lr0.50 2ap lar), femur II, ld(ep>0.80); metatarsus III, R4d(lp0,35 lep lr0.40 ler) R5v(lp0.40 lrO.35 lap lam lar) L4d(lp0.45 lep liU.40 ler) L5v(lp0.35 lrO.35 lap lam lar), tibia III, R4d(lp0.25 lpO.65 lrO.25 lr0.60) R3v(lr0.40 lap lar) L3d(lp0.65 lrO.35 lr<).65) L4v(lp0.40 lr0.40 lap lar), femur III, Rld(er); metatarsus IV, R3d(lr0.50 lep ler) R8v(lp0.15 lp0.50 lr0.15 lr0.50 lr0.60 lap lam lar) L4d(lp0.45 lrO.45 lep ler) L9v(lp0.30 lp0.50 lr0.15 lrO.45 lrO.55 lap lam 2ar), tibia IV, Rld(lr0.70) R4v(lp0.40 lrO.45 lap lar) L3d(li-0.25 lr0.70 lrO.85) L3v(lr0/35 lap lar); palpal tibia, Rld(p0.65) R3v(lp0.55 lp>0.80 lr0.40) Lld(p0.50) L3v(lp0.50 lp>0.80 lrO.45), pal- pal patella, ld(p0.60), palpal femur, Id(ep). Female: Allotype. Total length, 26.40; cara- pace, length, 9.00, width, 7.75, LCAVC =1.16; sternum, width, 4.05, length, 4.00; chelicerae, width, 4.80, length, 3.40. Cheliceral macroteeth. 8 each side, denticles, 19 right, 17 left. Maxil- lary cuspules, 101 right, 115 left; labial cuspu- les, 66. WChAVC = 0.62. General color black, carapace with a blue-green-black sheen, pubes- cence moderately dense, not appressed. Black pubescence of abdomen with dull green cast, slightly darker than carapace. Long, basally dark, pale orange-buff setae interspersed over entire dorsum and posterolateral surfaces, anter- odorsal setae with dark coloration extending fur- ther up shaft; anterolateral setae similar to an- terodorsal setae. Most ventral setae short and black, longer ventral setae sparsely interspersed and similar to anterolateral setae. Patch of black, type I urticating hair covering posterodorsal half of abdomen, not clearly visible because of dark color of pubescence and extensive interspersion of long orange-buff setae. Legs with black pu- bescence similar in color to abdomen; shortest setae mostly black with distal portion pale tan, longer setae pale orange-buff with basal portion black. Cephalic region slightly higher and rising more abruptly from thoracic region than in male; thoracic groove transverse, anterior mar- gin procurved; ocular turret occupying 23% of maximum cephalic width. AME circular, ap- proximately 3.5, AME- AME, 0.9 X AME di- ameter, AME- ALE, 0.4 X AME diameter, AME- PME, 0.3 X AME diameter; ALE ovoid, somewhat flattened on bottom, length, l.OX AME diameter, ALE-PLE, 0.6, 0.75 X AME di- ameter; PLE, right irregular, length, 0.7 X AME diameter, left subcircular, 0.6 X AME diameter, PLE-PME, 0.1, 0.15X AME diameter; PME elongate ovoid, length, 0.5 X AME diameter. Ventral setae of sternum as in holotype although homologous setae slightly less stout. Setae of ventral coxa and prolateral coxa I as in holotype. Femur HI not swollen. Leg and palp segment lengths are in Table 8. Extent of scopulae (X 100 = %): metatarsi I & II, to base (retrolateral scopula of metatarsus II, distal 0.80, medial scopula absent at very base); metatarsus HI, left, prolaterodistal 0.85 (0.50 dense), retrolateral 0.70 (0.55 dense), right, prolaterodistal 0.80 (0.65 dense), retrolateral 0.55 (0.45 dense); metatarsus LV, left, distal 0.40 medial, 0.25 pro- lateral, 0.20 retrolateral, right, distal 0.40 me- dial, 0.30 prolateral, 0.20 retrolateral. Tarsus IV scopula entire, not divided by setae; metatai*sus IV scopula, left, divided by setae proximal 40%, right, proximal 42 percent. Spination: metatai- sus I, Iv(am), tibia I, ld(p0.65) R5v(lr0.10 lr0.50 2ap lar) L4v(li*0.40 2ap lar), femur I, PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 159 Table 8. — Aphonopelma mojave new species, al- lotype female: leg and pedipalp segment lengths. I II III IV Palp Femur 7.05 6.45 5.90 7.30 5.20 Patella 3.80 3.50 3.10 3.40 2.85 Tibia 5.50 4.75 4.20 5.75 3.85 Metatarsus 4.75 4.65 5.20 7.10 Tarsus 3.70 3.60 3.60 4.00 3.60 Total length 24.80 22.95 22.00 27.55 15.50 Id(ep); metatarsus H, Rld(p0.35) Lld(p0.80) 3v(lr030 lap lam), tibia II, R2d(lp0.20 lp0.60) R5v(lr0.15 lrO.45 2ap lar) L2d(Ip0.25 lpO.65) L3v(lr0.41 2ap lar), femur H, Id(ep); metatarsus HI, R3d(lr0.40 lep ler) L4d(lp0.35 lr0.40 lep ler) 5v(lp0.40 lrO.35 lap lam lar), tibia m, R3d(lp0.60 lrO.45 lrO.85) L5d(lp0.20 lpO.55 lr0.15 lrO.55 lrO.85) 6v(lp0.50 lrO.15 lrO.35 2ap lar), femur HI, Rld(ep) L2d(lep lr0.60); metatarsus IV, 3d(lr0.45 lep ler) R8v(lp030 lp030 lr0.20 lr035 lrO.65 lap lam lar) L8v(lp0.50 Id). 15 li0.25 lr0.40 lrO.50 lap lam lar), tibia IV, 3d(lr0.15 lrO.65 li0.85) R6v(lr0.15 lrO.45 lrO.65 2ap lar) L5v(lp0.40 lrO.45 2ap lar), femur TV, Lld(er); palpal tibia, R2d(lp0.45 lpO.85) L2d(lp0.55 lpO.85) R8v(lp0.50 lpO.75 lrO.35 lrO.55 ler 2ap lar) L7v(lp0.20 lpO.55 lr0.40 ler2ap lar), palpal femur, Id(ep). Variation. — A. mojave consists of an east- ern and western race, geographically isolated from one another in the eastern and western Mojave Desert, respectively, by the Death Valley drainage. Eastern males have swollen third femora, and, although there is slight swelling in some western males there is no overlap in this character between the eastern and western races. The lower process of the tibial spur in eastern males is generally artic- ulated at a lesser angle and is apically more angular or curved than in western males. The smallest males of the species, found in Southern Utah, differ from other eastern males in having tibia I longer than metatarsus I in- stead of the usual reversed condition. While both conditions exist in western males there is quotient overlap only with eastern males from southern Utah. Eastern females have a relatively shorter metatarsus III than western females. In eastern specimens the carapace has a green-black or golden black (more com- mon) sheen, the abdomen and chelicerae a green-black cast; in western specimens the carapace has black or blue-black sheen, the abdomen and chelicerae a blue-black cast. Males: Total length, 15.1-21.1. Sternum length usually greater than but sometimes equal to width. Cheliceral macroteeth, eastern, 7-9, 7 most common (86%), 9 least common (2%), western, 6-9, 8 most common (55%), 7 (33%), 6 and 9 equally common (5%); den- ticles 4-20. Maxillary cuspules, western, 65- 130 (x = 101), eastern, 48-107 (x — 82); la- bial cuspules, western, 33-90 (x = 59), east- ern, 26-68 (x = 50). Long, pale orangish-buff setae of ventral abdomen may be sparse to moderately dense. Patch of type I urticating hairs covering distal 40-60% of abdominal dorsum. Tibia I arcuate, proximal bend mod- erate to strong, western males often with more pronounced bend. Ranges of leg and pedipalp segment lengths in Table 9. Little variation in morphology of the middle and apical portions of the palpal bulb (Figs. 16-21) and in form and articulation of basal division although sig- nificant variation in the shape of the basal di- vision. ALE, PLE, and PME vary consider- ably in relative size, PLE and PME in shape also; AME circular, most consistently shaped, AME-AME less than their diameter apart; ALE generally ovoid with ventral perimeter of eye somewhat flattened, in western specimens length usually equal to or slightly less than, seldom greater than AME diameter, in eastern Table 9. — Aphonopelma mojave new species, males (42 including holotype): range of leg and pedipalp segment lengths. I II III IV Palp Femur 6.40-9.35 6.00-8.80 5.55-8.10 6.40-9.60 3.75-5.60 Patella 2.95-4.45 2.80-4.20 2.50-3.70 2.70-3.95 2.10-3.05 Tibia 5.35-7.65 4.85-6.95 4.25-6.10 5.60-7.85 3.60-5.25 Metatarsus 5.25-7.70 5.30-7.65 5.75-8.40 7.15-10.50 Tarsus 3.70-5.10 3.60-5.10 3.50-5.20 3.80-5.80 1.60-2.40 160 THE JOURNAL OF ARACHNOLOGY Table 10. — Aphonopelma mojave new species, females (30 including allotype): range of leg and ped- ipalp segment lengths. I II III IV Palp Femur 4.55-7.85 4.00-7.20 3.55-6.65 4.55-8.15 3.35-5.75 Patella 2.55-4.60 2.30-3.90 1.95-3.35 2.30-3.65 1.90-3.05 Tibia 3.45-5.85 2.90-5.15 2.45-4.50 3.70-6.30 2.35-4.05 Metatarsus 2.95-5.30 2.80-5.20 3.00-5.70 4.15-7.85 Tarsus 2.40-4.00 2.30-4.00 2.30-4.10 2.60-4.50 2.30-4.00 specimens length usually greater than, seldom less than AME diameter; PLE usually smaller than ALE and larger than PME, usually ovoid, sometimes subcircular or irregular, PLE-PME barely removed or contiguous; PME often elongate ovoid or irregular, sometimes subcir- cular, generally smaller than PLE. Extent of scopulae (XlOO = %): retrolateral scopula of metatarsus II often short of base; metatarsus III, western, distal 0.60-0.85, eastern, 0.55- 0.85 (prolateral maximum); metatarsus IV, western, distal 0.30-0.45, x = 0.39, eastern, 0.25-0.55 (medial usually maximum), x == 0.33. Metatarsus IV scopula entire to divided, western, proximal 71%, eastern, proximal 86%. Ventroapical metatarsal spination: I, 1, II, 1-3 (usually 2), III, 2-6 (eastern, usually 4, western, usually 3), IV, 3-8 (eastern, usu- ally 4 or 5, western, usually 4). Females: Total length, 14.6-24,1. Sternum widest between coxae I and II; width greater than (in more than 50% of specimens), < length in western race; width less than length in eastern race. Cheliceral macroteeth, west, 7-9, 8 most common (63%), 9 least common, east 6-9, 7 most common (57%), 8 (33%), 9 least common. Maxillary cuspules each side, western, 79-149 (x = 111), eastern, 58-114 (x = 83); labial cuspules, western, 35-100 (x = 74), eastern, 26-85 (x = 54). Carapace pu- bescence usually not closely appressed al- though frequently more appressed than in males. Long orangish-buff abdominal setae varies in density, often more sparse ventrally than in males. Patch of type I urticating hairs covering distal 40-70% of abdominal dorsum. Ranges of leg and pedipalp segment lengths in Table 10. Spermathecae variable in both shape and distance between spermathecal bulbs (Figs. 24-28), variation not correlated with geography. Eye arrangement as in male, posterior eyes most variable in relative size and shape, AME least variable; distance be- tween adjacent eyes variable. Extent of scop- ulae (XlOO = %): metatarsi I and II as in allotype, metatarsus III, distal 0.55-0.85, western, x = 0.73, eastern, x = 0.62, meta- tarsus IV, distal 0.25-0.50 (western x = 0.38, eastern x = 0.32). Tarsus IV scopula entire, not divided by setae; metatarsus IV scopula usually partially divided by setae, less often, entire or completely divided. Ventroapical metatarsal spination: I, 1-2 (eastern, 1 or 2, western, usually 1) II, 1-3 (usually 2), III, 2- 6 (eastern, usually 4, western, usually 3), IV, 3-6 (eastern, most often 4, less often 5, west- ern, most often 4, less often 3). Distribution. — A. mojave is found through- out the Mojave Desert except in certain regions of south-central Nevada, areas that geographically isolate the eastern and western races and BDM populations from other east- ern populations, and most of JTNM. Only the northern-most populations near Goldfield Summit (western Nevada) inhabit biomes not characteristic of the Mojave Desert. East of Goldfield Summit toward the Beaver Dam Mountains, populations may exist in some of the less rocky valleys between the southern Nevada north-south mountain ranges although I am unaware of specimens collected in these areas. The Virgin Mountains, the rugged ter- rain toward the Grand Canyon, and decreasing elevations toward the Colorado River appar- ently bound the distribution of the eastern race. The (apparently) isolated populations of southern Utah and adjacent Arizona and Ne- vada were found as far south as the southern bajadas of the Mormon Mountains. The only known population in JTNM barely extends across the Monument’s boundary in the ex- treme northeastern comer, just west of the Coxcomb Mountains where Pinto Basin ex- ceeds 550 m elevation. A. mojave is consid- ered rare in the southwestern Mojave south of the diagonal from Queen Mountain (JTNM) PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 161 Figures 22-28, — Spermathecae of Aphonopelma Joshua new species and Aphonopelma mojave new species. 22, 23, A. Joshua, JTNM; 22, Covington Flats; 23, Fried Liver Wash, Pleasant Valley; 24, 25, A. mojave, west Mojave, Red Mountain, California; 26, A. mojave, west Mojave, NE Coxcomb Mountains, JTNM; 27, A. mojave, west Mojave, Yucca Valley, La Contenta Rd.; 28, A. mojave, east Mojave, Search- light, Nevada. to Pipes Canyon (foothills of the San Bernar- dino Mountains). The distribution of the spe- cies is shown on Map 1. Material examined. — Type specimens and the following: CALIFORNIA: San Bernardino Coun- ty: Halloran Summit, 0.5 mi. NW of L15, 4125 ft. elev., 19, 24 September 1989. Hwy 247, Rattle- snake Springs Rd. W of Johnson Valley, 3140 ft. elev., 19,1 November 1989. Apple Valley, S of A. V. toward Rattlesnake Mtn., 4060 ft. elev., 19, 4 November 1989. NE JTNM inside and outside of monument boundary, just west of Coxcomb Mtns. off Hwy 62, 12.2 mi. W of jet. Hwy 177, 2320 ft. elev., 19, 12 November 1989; 2050-2130 ft. elev., 3(3, 24 November 1989; 2050 ft. elev., 19,31 Jan- uary 1991; 2300 ft. elev., 19, 2 November 1991. Honda Rd., N of Yucca Valley off Hwy 247, 3890 ft. elev., 19, 14 April 1990; 3870 ft. elev., 19, 14 April 1991. Hwy 247, 20 mi. S of Barstow, 3200 ft. elev., IS, 16 October 1990 (S. Kutcher); IS, 26 October 1991. 7.5 mi. N of Pipes Canyon Rd. (Pipes Wash), 3680 ft. elev., definitive molt, late September-early October 1991; lc3, 14 April 1991. 3.7 mi. N of Pipes Canyon Rd., 3400 ft, elev., 1 9, 18 April 1991. E of Yucca Valley, La Contenta Rd. between Hwy 62 and Yucca Trail Rd., 3275 ft. elev., 19,2 November 1991. Apple Valley, 2 mi. S on Milpas Dr. off Hwy 78, 3040 ft. elev., 19,5 May 1992. Cima, N of Cima, 4.6 mi. W on power line road, 4350 ft. elev., IS, 10 October 1992. 2.9 mi. W on power line road, 4590 ft. elev., 29, 11 October 1992. Black Canyon Rd., 2.8 mi. N of Es- sex Rd. jet, 3240 ft elev., 19,11 October 1992; 19, 25 October 1992. Kelbaker Rd., 9.4 mi. S of Kelso, 3120 ft. elev., 1(3, 31 October 1992. Cima, 1.5 mi. W of Kelso-Cima Rd. W of Cedar Canyon Rd., 3800 ft. elev., 1(3, 31 October 1992. Kelbaker Rd., 8 mi. S of Kelso, 2880 ft. elev., 1(3, 1 Novem- ber 1992. Kelso-Cima Rd., 10 mi. N of Kelso, 3300 ft. elev., 1(3, 1 November 1992. Morning Star Mine Rd., 3.1 mi. SW of Ivanpah Rd. 3005 ft. elev., Ic3, 1 November 1992. Nipton, between Nipton and Ne- vada state line, 3245 ft., 3705 ft. elev., 2S, 1 No- vember 1992. Kern County: Hwy 395, 8.1 mi. N into Kern Co., 3500 ft. elev., Ic3, 20 October 1991. Inyo County: Death Valley National Monument, jet. Harrisburg Flats, Skidoo Rd., 5000 ft. elev., 2(3, 18 October 1963 (R. Hardy). Los Angeles County: San Gabriel Mtns. foothills, north slope off N2, 1(3, 21 October 1976 (M.E. Thompson). Canyon Country, N on Hwy 14, IS, 30 October 1978 (M. Wilker- son). Valyermo, Bob’s Gap Rd., 1.5 mi. N of N4, 4050 ft. elev., 1(3, 28 October 1989. NEVADA: Clark County: Searchlight, 0.5-3. 0 mi. W of SL., 3300 ft. elev., 4(3, 23 October 1976 (W. Icenogle). 8.2 mi. W of SL., 0.5-1.5 mi. N of Hwy 164, 4280 ft. elev., 29, 7 October 1989; 4365 ft. elev., 19, 12 October 1990; 4180-4260 ft. elev., 3 9, 12 Oc- tober 1991. Nye County: 10 mi. W of Mercury, IS , 3 November 1972 (W. Icenogle). Scotty’s Jet., 10 mi. S on Hwy 95, 4000 ft. elev., 2(3, 28 October 1978 (W Icenogle). Lida (Hwys 266 and 95 jet.), 5.5 mi. S, 4500 ft. elev., 1(3, 28 October 1978 (W. Icenogle). Esmeralda County: Goldfield Summit, 8 mi. S on Hwy 95, 5000 ft. elev., 1(3, 28 October 1978 (W. Icenogle). UTAH: Washington County: Beaver Dam Mtns., Summit Springs off old Hwy 91, 4140 ft. elev., 19, 6 October 1993; 3960-4120 ft. elev., 3 9, 12 October 1993. Old Hwy 91, 2.7- 3.5 mi. N of Utah- Arizona line, 3140-3300 ft. elev., 162 THE JOURNAL OF ARACHNOLOGY Map 2. — Distribution of Aphonopelma iodium. The boundaries of the Mojave Desert (as percieved by the author) are indicated by the outer-most dot- ted lines. 4d, 19-20 October 1993. W of Hwy 91, 2.4 mi. W of Welcome Springs Rd. turnoff, 3680 ft. elev., 1 S . Specimens collected by the author deposited in AMNH. Aphonopelma iodium (Chamberlin & Ivie) Figs. 9, 10, 29-50; Map 2 Delopelma iodius Chamberlin & Ivie 1939: fig. 3 (male holotype from Washington County, Utah, 2 miles west of Castle Cliffs (Beaver Dam Moun- tains), 27 November 1936, in AMNH, examined). Aphonopelma iodius: Chamberlin 1940: 7. Aphonopelma iodium: Smith 1994: 115. Spelling change, gender neuter. Delopelma melanius Chamberlin & Ivie 1939: fig. 1 (male holotype from Salt Lake County, Utah, University of Utah campus, September 1925, in AMNH, examined; female allotype lost). Aphonopelma melanius: Chamberlin 1940: 6. NEW SYNONYMY. Aphonopelma melanium: Smith 1994: 120. Spelling change, gender neuter. Aphonopelma nevadanum Chamberlin 1940: 13 (male holotype from Clark County, Nevada, col- lected by G. Carter, searchlight, 2 December 1930, in AMNH, examined). NEW SYNONY- MY Aphonopelma angusi Chamberlin 1940: 21—22 (male holotype and female allotype from Wash- ington County, Utah, collected by A.M. Wood- bury, R. Hardy, H. Higgins, and R. Pendleton, 2 miles west of Beaver Dam Mountains, 7 October 1939, in AMNH, examined). NEW SYNONY- MY Synonymy. — Aphonopelma melanium, A. angusi, and A. nevadanum are placed in the synonymy of A. iodium (one of two possible senior synonyms) with which they share all characters of specific significance (Tables 1, 2); there are no other characters known that merit their continued separation. The A. an- gusi allotype has a shorter carapace (carapace length 9.10) than any other conspecific female examined in this study; the ratio of its cara- pace length to that of a larger female from the type locality is 0.64. However, corresponding ratios of the smallest to largest females in A. Joshua and A. mojave and males in A. iodium are 0.64, 0.59, and 0.66, respectively (smallest and largest of each species from the same lo- cality). Metatarsi I and II are proportionately shorter relative to femur I in the A. angusi allotype than in other A. iodium females. Al- though no other correlation between the size and proportional leg or leg segment length was found within the combined sample, the proportionately shortest metatarsi I and II (ex- cluding A. angusi allotype female) were found in the smallest female (carapace length 10.80) and the proportionately longest metatarsi in the largest female (carapace length 22.05). Therefore, both carapace and relative metatar- sal lengths in the allotype are believed to be extensions of the female range for these char- acters. Leg and pedipalp segment lengths of the A. angusi allotype are in Table 13. All "eutylenum type’ tarantulas of the Mojave Desert are considered A. iodium, sharing with the type all specifically significant characters. Ecological, behavioral, and distribution data gathered from this assemblage support the synonymy of A. iodium. Diagnosis. — Aphonopelma iodium is easily distinguished from A. Joshua, A. mojave, and A. steindachneri by extensive scopula of metatarsus IV and by the pale-buff color of the carapace and of the patella and tibiae of legs I and II in females. There are only two valid species Ceutylenurn types’?) described PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 163 prior to 1939, A. rusticum (Simon) and A. hel- luo (Simon), in which the carapace coloration (as described) and extent of metatarsus IV scopula are similar to the corresponding char- acters in A. iodium (ambiguity in proper type representation of A. rusticum is discussed in the ‘Status of some old Eurypelma species’ subsection above). A. iodium is distinguished from A. rusticum (USNM, cotype #1585) by longer legs relative to carapace length (Table 1); although most segments of leg I of the cotype are missing, LFI/LC = 0.88; in A. io- dium LFI/LC = 0.93-1.07. The A. rusticum paratype specimen (MNHP-Paris, #5873), considered the lectotype by Smith, is doubt- fully of the same species as the USNM cotype specimen since the length of the patella plus tibia IV (also patella plus tibia I) is less than the length of the carapace in the former (LP -I- TIV/LC < 1.00) and greater than the car- apace length in the latter (LP + TIV/LC = 1.11); in A. iodium males LP + TIV/LC = 1.18-1.32 which clearly distinguishes it from both specimens. Proportionately longer legs in A. iodium males further distinguishes the spe- cies from that of the Mazatlan paratype male. Based on Simon’s locality data, I believe that A. rusticum is most likely a summer breeder. A. iodium is distinguished from A. helluo (ho- lotype male #17707 and non-type male #50(44)) by a shorter carapace and longer legs relative to carapace length and again from the non-type male by more extensive scopulae of metatarsi III and IV and proportionately lon- ger metatarsi I and II (measurements of the holotype were taken from Smith; numerical and character data from the non-type male are in Table 1). Other species in the 'eutylenum group’ include A. eutylenum, A. clarum, A. brunnium (brunnius), A. cryptethum (crypte- thus), A. cratium (cratius), A. prosoicum (pro- soicus), and A. griseum. Although A. iodium can be distinguished from all of these types by various leg and palpal length proportions (or segment proportions), the character differ- ences in the types of A. eutylenum, A. clarum, A. brunnium, A. cryptethum, and A. cratium appear to be minimal. Consequently, the un- ambiguous separation of A. iodium from any of the five species will remain questionable until the variational limits of these quantita- tive characters have been determined for var- ious populations of inland and coastal 'eu- tylenum types’. Description. — Males: Carapace, length 9.35-16.90 (x = 13.0), width 8.10-15.60; smallest males found in southern Utah. Ster- num, length 4.20-7.60, width 3.80-6.85; usu- ally longer than wide, length equal to width in one male from JTNM. Chelicerae, width 4.20-7.95. Cheliceral macroteeth 7-9, denti- cles 4-16. Labial cuspules 55-140 (x = 102); maxillary cuspules 103-234 (each side) (x = 181). Color of carapace pubescence pale buff or paper-bag brown (usually darker with greenish-bronze sheen following molt, most pronounced in cephalic region); black to dark brown abdomen and appendages; chelicerae similar in color to carapace but usually slight- ly darker. Patch of black type I urticating hairs covering posterodorsal 45-65% of abdomen, difficult to distinguish because of pubescence coloration and interspersion of long orange- tan setae. Abdominal anterodorsal setae spi- niform, stout, and uniformly black or dark reddish-brown; anterolateral setae also dark but shorter and more slender than dorsal setae; longest setae filiform, orange-tan with dark basal portions, interspersed on posterodorsal, posterolateral, and caudal surfaces, the longest inside and just outside of patch of urticating hair, the shortest toward ventral margins; ab- dominal venter usually with sparse intersper- sion of similar setae and a dense covering of short, fine, dark setae. Sternum with relatively slender, attenuate medial setae and more stout, spiniform marginal setae; setae intermediate in position also intermediate in basal diameter. Coxae (LIV) with retromarginal, promarginal, and distal setae similar to marginal sternal se- tae; most basomarginal and medial setae sim- ilar to intermediate and medial setae of ster- num. Baso- and retromarginal setae of palpal coxae similar to medial or intermediate sternal setae. Sternal and ventral setae of coxae very similar to, if not indistinguishable from, ho- mologous setae of A. mojave males (Figs. 3, 4, respectively). Setae on prolateral surface of coxa I spiniform and basally swollen (Figs. 9, 10). Leg setae attenuate, mostly pale buff with dark basal portion, the shortest mostly dark with pale distal ends. Metatarsus IV almost always longer than length of carapace, rarely equal to and always longer than femur I; metatarsus I generally longer than tibia I but can be slightly shorter than tibia I in males from southern Utah. Leg and pedipalp seg- ment lengths of the holotype are in Table 1 1 ; 164 THE JOURNAL OF ARACHNOLOGY Table 1 1 . — Aphonopelma iodium, holotype male: leg and pedipalp segment lengths. I II III IV Palp Femur 15.40 14.60 13.35 15.20 9.15 Patella 7.10 6.60 5.80 6.20 4.75 Tibia 12.60 11.55 9.90 12.40 8.30 Metatarsus 13.35 13.00 13.60 16.90 Tarsus 8.10 7.90 7.60 8.40 3.30 Total length 56.55 53.65 50.25 59.10 25.50 ranges of segment lengths in Table 12. Retro- lateral bend into apical division of the palpal bulb uniform rather than abrupt and bulb rel- atively slender (Figs. 29, 31, 33, 35, 37, 39, 41, 43) compared to bulb of A. mojave\ prox- imal prolateral protuberance on the dorsal as- pect prominent (Figs. 30, 32, 34, 36, 38, 40, 42, 44) as in A. mojave males. Extent of scop- ulae (XlOO = %): metatarsi I and II to base, metatarsus III, prolateral, usually to base or near to base, medial, 0.60-0.80, metatarsus IV, retrolateral, distal 0.70-0.85, medial, 0.40- 0.55. Tarsal scopulae entire, not divided by setae. Ventroapical metatarsal spination: I, 1- 3 (usually 2), II, 1-4 (usually 3), III, 2-5 (of- ten 4, less often 3), IV, 2-5 (almost always 4). Females: Carapace, length, 10.80-22.05 (x = 15.02), width, 9.65-18.70. Sternum, length, 4.60-8.70, width, 4.70-7.80, usually longer than wide but slightly wider than long in smallest female. Chelicerae, width, 6.25- 12.65. Cheliceral macroteeth, 7-10, denticles, 7-15 (x ^ 11). Labial cuspules, 100-138 (x = 114); maxillary cuspules 129-267 (each side), X = 195. Color of carapace and chelic- erae as in males; color of tibiae and patellae of legs I, II, and palps similar to carapace (col- or varying in degree), corresponding segments of legs III & IV less accentuated but usually slightly lighter than the remaining leg seg- ments. Patch of black type I urticating hairs covering posterior Vi-^A of abdominal dorsum. Abdominal anterodorsal setae similar to those of males but less stout and distally pale buff; longest setae basally dark, orange-tan, distrib- uted as in males relative to patch of urticating hair but often dispersed further anteriad. Ster- nal, coxal, and leg setae as in males although homologous setae often slightly less stout. Metatarsus IV and femur I always shorter than length of carapace; metatarsus I usually short- er than tibia I but longer than in largest female (JTNM); femur IV longer than femur I; meta- tarsus III longer than metatarsus I. Ranges of leg and palpal segment lengths are in Table 14. Variations in the shape of spermathecae and relative distance between bulbs (Figs. 45- 50) are inconsistent with population geogra- phy (it is not cuiTently known if spermathecal characters adequately distinguish any Aphon- opelma species). Extent of scopulae (XlOO = %): metatarsi I and II, to base, metatarsus III, lateral, to base or close to base, medial, distal 0.70-0.80, metatarsus IV, retrolateral, 0.75- 0.85, medial, 0.50-0.60. Tarsal scopulae en- tire, not divided by setae. Ventroapical meta- tarsal spination: I, 1-3 (usually 2), II, 1-3 (of- ten 3, less often 2), III, 3-4 (equally common), IV, 4. Distribution. — A. iodium is common throughout the Mojave Desert west of the Col- orado River, its distribution continuous to the north into the Great Basin in Utah and Ne- vada. Its distribution to the south and to the west of the Mojave Desert and to the north- ern-most limits in Nevada and Utah has not yet been determined. However, preliminary data from extensive fieldwork suggest that to the south (excluding the low desert) and west A. iodium is replaced by an inland and coastal species (‘eutylenum type’) while to the north it is the only theraphosid species, other than A. mojave, found in Nevada and the only 'eu- tylenum type’ found in Utah. The known dis- tribution of the species within the Mojave Desert is shown on Map 2. Table 12. — Aphonopelma iodium, males (32): range of leg and pedipalp segment lengths. I II III IV Palp Femur 9.70-17.55 9.20-16.80 8.50-15.50 9.65-17.70 6.00-10.45 Patella 4.70-8.25 4.25-7.60 3.70-6.85 3.95-7.20 3.30-5.50 Tibia 8.15-13.75 7.15-12.90 6.15-11.45 7.85-13.80 5.40-9.95 Metatarsus 8.15-16.10 8.10-15.75 8.45-16.35 10.75-20.60 Tarsus 5.30-9.30 5.10-8.80 5.00-8.80 5.70-9.70 2.45-3.80 PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 165 Figures 29-44. — Palpal bulbs of Aphonopelma iodium, right; odd, ventral; even, dorsal (arrow-proximal prolateral protuberance). 29, 30, Holotype, A. iodium; 31, 32, Holotype, A. angusi (left bulb); 33, 34, Holotype, A. melanium; 35, 36, Holotype, A. nevadanum (tip of embolus broken); 37-44, Mojave Desert; 37, 38, Mojave Desert, BDM, Utah; 39, 40, Mojave Desert, Searchlight, Nevada; 41, 42, Mojave Desert, Quail Mountain, JTNM; 43, 44, Mojave Desert, Red Mountain, California. Specimens examined. — Holotype male, holo- types: A. melanium, A. angusi, and A. nevadanum, allotype: A. angusi, and the following: CALIFOR- NIA: Riverside County: JTNM: Cottonwood Springs, 3000 ft. elev., definitive molt, 21-22 Sep- tember 1986; Id, 25 February 1985. 1.0-1.5 mi. E of Cottonwood Springs, 3300 ft. elev., 19,4 April 1989. Hexie Mtns. W of Cholla Cactus Gardens on Pinto Basin Rd., 2760 ft. elev., 1 9, 16 April 1989. Quail Mtn., 5.7 mi. SE of monument entrance off Table 13. — Aphonopelma angusi, allotype female (=A. iodium): leg and pedipalp segment lengths. I II III IV Palp Femur 7.35 6.70 6.15 7.55 5.40 Patella 4.00 3.65 3.30 3.60 3.10 Tibia 5.55 4.70 4.15 5.90 4.00 Metatarsus 4.85 4.70 5.20 7.30 Tarsus 4.00 3.80 3.90 4.50 4.10 Total length 25.75 23.55 22.70 28.85 16.60 Quail Mtn. Rd., 0.5 mi. SW of picnic area, 4080 ft. elev., 19, 5 August 1989. San Bernardino Co.: Reche Rd., 4.5 mi. E of Landers, 2950 ft. elev.. Id, 18 October 1981 (W. Icenogle). NE Coxcomb Mtns., 8 mi. W of jet. Hwy. 177 on Hwy 62, 1600 ft. elev., 2 9, 4 February 1990. East Mojave, Mid Hills toward campground, 5700 ft. elev., definitive molt, 2 August 1989; Id, 13 May 1989. JTNM: Quail Springs Rd., 2.7 mi. SE of monument en- trance, 4000 ft. elev.. Id, 3 August 1989. Coving- ton Flats, 0.4 mi. N of Monument boundary, 4220 ft. elev., Id, 10 August 1989. Kelso-Cima Rd., 6.3- 11.5 mi. N of Kelso, 2820-3480 ft. elev., 2d, 1 November 1992. Hwy 247, 10 mi. S of 1-15 at Bar- stow, 2840 ft. elev.. Id, 7 November 1992. Kramer Jet., 3.1 mi. E on Hwy 58, 2470 ft. elev.. Id, 8 November 1992. San Bernardino and Kern Co. lines (Red Mtn. area): 19-23 mi. N of Kramer Jet. on Hwy 395, 0.5-1. 5 mi. W of highway, 3220- 3400 ft. elev., 2d, 20 October 1989; 3200-3280 ft. elev., 49, 12-14 October 1991; 3d, 26 October 1991; Id 19, 13 October 1992; 19, 22 January 1994. Kern County: California City, 7.6 mi. E on 166 THE JOURNAL OF ARACHNOLOGY Table 14. — Aphonopelma iodium, females (14): range of leg and pedipalp segment lengths. I II III IV Palp Femur 9.10-17.00 8.40-15.85 7.90-14.60 9.65-17.30 6.90-12.50 Patella 4.95-9.10 4.55-8.60 4.15-7.85 4.55-8.30 3.65-6.80 Tibia 6.80-12.40 5.90-11.20 5.35-10.00 7.30-12.40 4.90-9.00 Metatarsus 6.20-13.30 6.00-12.80 6.65-13.70 9.05-17.95 Tarsus 4.70-8.80 4.50-8.60 4.70-8.60 5.10-9.30 5.00-9.00 Twenty Mule Team Rd., 2710 ft. elev., Id, 20 Oc- tober 1991. Inyo County: Deep Springs Valley, ap- prox. 2 mi. E of Westgard Pass on Hwy 168, ap- prox. 5000 ft. elev.. Id, 16 October 1976 (Frank Hovore); Big Pine, 2.3 mi. E on Hwy 168, 4320 ft. elev.. Id, 20 October 1993. NEVADA: Clark County: Pahmmp, 17-19 mi. SE on Hwy 160, 3400 ft. elev.. Id, 11 October 1974 (W. Icenogle). Searchlight, 3 mi. S on Hwy 95, 3250 ft. elev.. Id, 2 October 1981 (W. Icenogle). 8.2 mi. W on Hwy 164, 2 mi. N to foothills of Highland Range, 4330 ft. elev., Id2?, 7 October 1989. UTAH-ARIZO- NA: Washington County, UTAH and Mohave County, ARIZONA: Male specimens collected be- tween 2320-4330 ft. elev. in Castle Cliffs, Summit Springs, and Welcome Springs areas on southem slope of Beaver Dam Mountains and close to Utah- Arizona border on both sides of state line off old Hwy 91, limmd, 4 July 1989; definitive molt, 8 September 1989; Id, 16 September 1989; Id, 23 September 1989; 3d, 6 October 1993; Id, 12 Oc- tober 1993; 5d, 19 October 1993; Id, 20 October 1993. Female specimen collected in Summit Springs area in Beaver Dam Mountains, 3650 ft. elev., 1$, 5 October 1993. All specimens collected by the author unless otherwise indicated. Speci- mens collected by the author deposited in AMNH. NATURAL HISTORY Habitat. — A. Joshua new species and A. mojave new species, only narrowly sympatric (near Yucca Valley, California), appear to have similar ecological requirements through- out their ranges. Both burrow in soil of like composition, inhabit comparable vegetation communities, and occur at elevations between 550-1600 m. A. mojave has been found at slightly lower elevations near Trona, Califor- nia, and A. Joshua also inhabits biomes char- acteristic of the Sonoran Desert (southeastern JTNM). Both species prefer flat or gently sloping terrain composed of sandy soil of var- ious particle size. A. iodium, sympatric with both A. Joshua and A. mojave throughout their respective ranges, occurs at elevations below 300 m in the basins and above 1700 m on the Figures 45-50. — Spermathecae of Aphonopelma iodium. 45, A. angusi allotype (—A. iodium)’, 46, Searchlight, Nevada; 47, 48, Red Mountain, California; 49, Quail Mountain, JTNM; 50, Pleasant Valley, JTNM. PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 167 inter-desert and perimeter mountain slopes and is equally common on steeper and/or rockier hillslopes and in regions where the substrate is rich in clay content. Because of the increased tolerances of A. iodium to varied substrate consistencies, xeric conditions, and cooler, mildly hydric climates, it has succeed- ed in accessing regions that are unsuitable for A. mojave, both within the Mojave and to the north, in the Great Basin Desert. The xeric drainages that divide A. mojave into eastern and western races, the rocky, rugged terrain between Las Vegas and the Morman Moun- tains that apparently isolates northeastern A. mojave populations, and the mountainous regions to the north of St. George were not found to be geographic barriers for A. iodium (a corridor, 1700 m in elevation, snakes through Utah from the BDM to the Idaho bor- der (Salt Lake City to Provo to Nephi, west to Jericho, south to Delta, to Black Rock, to Milford, to Lund, east to Cedar City, and fi- nally south to St. George between the Pine Valley Mountains and the Hurricane Cliffs); two specimens from Utah and one from Cal- ifornia were found at elevations greater than 1670 m. All Mojave species are common in vege- tation communities dominated by Creosote Bush (Larrea), often in pure stands or in as- sociation with either bursage {Ambrosia), brit- tlebush (Encelia), or Joshua-tree (Yucca), and by Joshua-tree and perennial bunchgrass in as- sociation with Creosote Bush and/or sparsely distributed blackbrush (Coleogyne). Commu- nities within which A. iodium is relatively common (but not usually abundant) but A. Joshua and A. mojave rare or absent, are dom- inated by saltbrush or bursage at low eleva- tions (below 600 m) and by Blackbrush, Big Sagebrush (Art erne sia), and juniper or mixed juniper and pinon at high elevations (above 1500 m). Daily activity patterns. — After initially unplugging their burrows following the cool winter months, the Mojave Aphonopelma de- posit silk around entrance perimeters to a greater or lesser extent; and over time many layers of silk accumulate. During daylight hours of the ensuing months most burrows have a sheet-like layer of silk (common in all three species) or a loose plug of silk-bound earth (less common and only in A. Joshua and A. mojave) blocking the entrance (obstruction of the entrance may ward off certain potential day-time predators such as ants and help maintain higher humidity in the burrow). Usu- ally after dark or, less frequently, just before sunset when overcast or cloudy, females and immatures were commonly seen just inside their burrows or a few centimeters below the entrance. Only after dark (all observations) were the silken coverings and plug removed, the silken sheets torn down and flattened against the perimeter and inside wall of the burrow, the plugs either pushed with forelegs or carried in the chelicerae (with the aid of the palps) to the outside of the burrow. After re- moval of these obstructions the majority of spiders, initially, remained motionless within the burrow confines, completely submerged in the upper level or with forelegs resting over the perimeter. After varying intervals of time many individuals then exited their burrows, eventually to assume a waiting position. De- position of additional silk around the entrance perimeter was one of the first activities of the resident, once above ground, followed by (rarely preceded by) intermittent, brief periods of wandering, between which the waiting pos- ture was assumed. A. Joshua and A. mojave rarely ventured more than several cm from the entrance (except when in pursuit of prey) al- though individuals of both species were ob- served at distances of up to 30 cm. The larger A. iodium were commonly observed at dis- tances of 30 cm or less and, rarely, of up to one meter. Seasonal activity. — Seasonal activity is de- fined by the period of time between which burrows of a given species are first unplugged after winter and then replugged for the follow- ing winter. The presence of fresh silk around the entrance perimeter, recently excavated ma- terial around the burrow, or a thin sheet of silk covering the burrow’s entrance were indica- tive of desert Aphonopelma activity. Con- versely, an open burrow lacking silk around the entrance, accumulated dust on the perim- eter silk, or a burrow with a hardened plug or turret were indicative of inactivity. Active burrows of A. Joshua were found be- tween the end of March and the last week of October, those of A. mojave between the first week of April and the first week of January, and those of A. iodium throughout the year. In A. Joshua the vast majority of burrows were plugged by early October but in A. mojave not 168 THE JOURNAL OF ARACHNOLOGY Figure 51. — Burrow of Aphonopelma Joshua new species (female) showing typical turret made by the species; the shadow cast by the turret gives an indication of its height (A. mojave new species turrets are indistinguishable from those of A. Josh- ua new species). until middle or late November. Activity within A. Joshua was abundant by mid-spring and through the species summer breeding season but steadily declined thereafter. In contrast, activity within A. mojave was minimal until close to the beginning of the species fall breeding season, peaking by mid-October; winter plugging coincided with the termina- tion of the breeding season. A. iodium was commonly found between early March and mid-December, but activity was most abun- dant in late summer and during the species fall breeding season. Open burrows were infre- quently discovered in January and February, and then, only in the central and southern por- tions of the species range. Usually prior to or during respective breed- ing seasons, excavations composed of silk- bound soil, sclerotized remnants of prey, and old exuviae were deposited outside of the bur- rows in a species dependent fashion; only a few individuals excavated (any material) in spring. Both A. Joshua and A. mojave formed indistinguishable turrets (Fig. 5 1 ) with the ex- cavations that surrounded and elevated the en- trance (the vast majority of turrets formed within a given year were washed away by the hard winter rains leaving no trace of the bur- row’s location; the few that were not de- stroyed had become hardened mounds that were built upon with new excavations prior to or during the subsequent breeding season). Turrets as high as 13 cm were found although most were less than 6.5 cm, with the average outside diameters approximately 5-6 cm. The inside turret walls were usually lined with silk which continued over the entrance and par- tially or completely blanketed the top of the mound. A. iodium, on the other hand, scattered excavations loosely around (rarely, appearing slightly mounded) or to one side of the bur- row, often at some distance from the entrance; excavations were infrequently found by bur- rows known to contain medium to large size females. Entrance perimeters were generally lined with copious layers of silk which ex- tended up to several centimeter both over the substrate and into the burrow; the entrance al- most always opened at substrate level. Exca- vation by all species, although often intermit- tent, generally appeared to be a continuous process throughout respective breeding sea- sons. A proportion of each species replugged their burrows on a routine (observed only dur- ing a breeding season), irregular (for relatively short durations), or extended period (observed for up to two months) basis during their sea- sonally active months. Minch (1979) reported both intermittent and extending plugging by A. chalcodes Chamberlin. In the Red Moun- tain area several A. mojave burrows were reg- ularly plugged (from one to three days ob- served) and reopened (only one night observed) during the fall breeding season; plugs were usually in place well before sun- rise and pushed aside only after dark. Other bun'ows, previously not plugged but covered by a sheet of silk during the day, were plugged in late September and early October, reopened in late October toward the end of the breeding season, and shortly thereafter, replugged for the winter. One burrow, found plugged at the beginning of the breeding season, appeared to have been recently active because of its only slightly hardened turret (turrets that were plugged or otherwise inactive for long periods of time developed a windblown, smoothed ap- pearance and very hardened outside walls; new silk and excavation were lacking); and its occupant removed the plug and began to ex- cavate in the last week of October. Summer plugging (Minch 1979) by A. mojave was ob- served only in captives, several of which plugged their burrows prior to a subsequent summer molt, one prior to producing an egg sac; several others simply sealed the entrance PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 169 with silk before molting. Since gravid females and females with egg sacs were rarely found during the summer months (and then, in bur- rows with only silken sheets covering the en- trance), I suspect that most do not become seasonally active until their young are ready to disperse. Summer plugging by A. iodium was not uncommon but burrows were rarely found plugged during the species breeding season. One burrow (in JTNM) plugged be- fore 5 August was reopened between late Sep- tember and 12 October and remained open for the duration of the breeding season. Minch (1979) reported that the maximum duration of winter plugging by A. chalcodes (female) was between 644-674 consecutive days. One immature and two female A. palo- ma were taken from inactive burrows (bur- rows were not detectable other than by my markings and had been monitored monthly) on 19 November 1992 (near Sentinel, Arizo- na), at least 711 days after they were plugged prior to 9 December 1990. Although these data are lacking for the Mojave Aphonopelma, burrows (and wandering males) of all species were abundant throughout the Mojave in 1989 but were difficult to locate in 1990. Again in 1991 burrows were abundant, found in even greater numbers than in 1989. Yearly burrow density fluctuations at all study sites (for each species) followed this larger scale trend. Such observations lead me to believe that extended plugging within desert species may be similar in duration to that observed in A. chalcodes and A. paloma. Burrow construction and remodeling. — Although all observed Aphonopelma species can construct their own burrows, evidence in- dicates that most individuals that have aban- doned or been displaced from their burrows will adapt any suitable cavity. In captive sit- uations intentional burrow damage by the in- vestigator initiated new burrow construction by the former occupant only if no other cav- ities were available. In similar field experi- ments, a ‘homeless’ spider sought shelter in the first sufficiently large unoccupied burrow or subterranean cavity encountered. Recent occupancy was suggested when a burrow was found with unplugged side tunnels or was sig- nificantly wider in diameter than usual for the size of the resident. Whether growing taran- tula spiderlings (under natural conditions) continue to enlarge and remodel initial bur- rows or seek larger burrows as needed is not known although all Aphonopelma I have had in captivity continued to utilize initial burrows through their frequent enlargement. Most desert captives eventually plugged pre-existing burrows only to resurface either back through the plugs or through new shafts. Entrance and upper burrow diameters were consequentially reduced and in the majority of burrows were correlated with occupant size. Natural entrance diameters for A. iodium ranged from approximately 1 .2-2 times great- er than the width of the carapace and those for A. Joshua and A. mojave from 1,5-2 times greater. Minimal diameters may prevent intru- sion by slightly larger, more powerful preda- tors or conspecifics, lessen the effects of ero- sion, especially when burrows are reopened through a plug, and help to maintain optimal humidity within the burrow. Burrows of A. Joshua and A. mojave ex- tended more or less vertically to depths be- tween 25-53 cm. Well established burrows (those with side chambers packed with dis- carded food remains and pieces of old exu- viae) were usually the deepest. Throughout most of their length typical burrows were larg- er in diameter than at subsurface and entrance levels although were commonly constricted in up to several regions, one of which was usu- ally near or adjacent to the horizontally in- clined terminal chamber. Side tunnels and shafts beyond certain depths were usually plugged with silk-bound earth and shallow burrows were presumed to have been exca- vated to appropriate depths based on the vary- ing quantities of excavated material found outside the entrances (other than during breed- ing seasons). The average depth of A. iodium burrows was approximately 45 cm, ranging from 30 cm to 1 m. Their tortuous burrows commonly ended in horizontal chambers sim- ilar to those of A. Joshua and A. mojave. Short side chambers in well established burrows were generally located near the bottom and were used as dumpsites for accumulating food debris, old exuviae, and discarded egg sacs. Molting cycles. — Molting cycles are known primarily from captive specimens (ob- servation over several years for most juve- niles), Immature and adult female A. mojave (16 adult) generally molted between mid- July and mid-August. However, one immature fe- male molted as early 22 May and two females 170 THE JOURNAL OF ARACHNOLOGY molted after 1 1 August, one immature and one adult on 7 and 9 September, respectively. De= finitive molts of three males occurred between late August and the first week of October. Im= mature male cycles were coordinated with fe- male cycles. In A. Joshua, penultimate males collected within 7-8 months of maturity molt- ed between 7 July-1 August (just prior to or during their breeding season). Subadult males collected within two or three molts of maturity molted as early as 24 June and as late as 1 September. Juveniles (except those molting more than once a year) and females molted between 8 June-7 August. In A. iodium ju- veniles and females were known to molt be- tween late May and early October, the major- ity molting in July. One immature female was found in its natural burrow 23 September 1989 (BDM) with not yet fully sclerotized fangs, indicating a very recent molt. Two cap- tive males (penultimate instars when collec- ted), one from the BDM (Utah) and one from the Mid Hills area (east Mojave), molted 8 September and 2 August, respectively. Both Minch (1979) and Baerg (1958) found molting to be predominantly an annual event except in rapidly growing spiderlings and young juveniles (Baerg and Minch), in fe- males producing eggs (Minch), and in older females (Baerg). However, a proportion of ju- veniles and subadults and an even larger pro- portion of adult females in captive Mojave Aphonopelma failed to molt annually. Under natural conditions it was not uncommon to find unusually faded specimens, a condition indicative of a skipped molt. Old females (age estimated by relative size) were infrequently found not only with pubescence very bleached but worn away in various areas on the legs, carapace, and chelicerae. Such females may have weathered two or more consecutive years without molting. Baerg observed that Arkansas females (A. hentzi) producing egg sacs delayed molting until shortly after dis- persal of the young. Minch noted that female A. chalcodes producing egg sacs failed to molt in that same year. In agreement with Minch, I observed that captive females (A. Joshua and A. mojave) subsequently failed to molt in the year they produced egg sacs but molted the following year. Similarly, females (A. Joshua- 1, A. moJave-3, A. iodium-l) taken from the field with egg sacs failed to molt until the fol- lowing year. When bunows were unearthed during or shortly after a breeding season, re- cently discarded egg sacs were not found with females appearing to have molted within the year; coloration of females was faded when such egg sacs were found, indicating skipped molts. Apparently most Mojave Desert fe- males are not able to acquire the nutritional reserves necessary for egg production and subsequent molting activity. Sperm webs. — Sperm webs of the desert species are typical of those described by Baerg (1958) of A. hentzi. In using his de- tailed accounts of male behavior during web construction as a comparative reference I found no distinguishable differences between conesponding behaviors of A. hentzi and the Mojave species (captive males). Initial webs of A. Joshua males were con- stmcted inside the burrow, if sufficiently wide in any region or, otherwise, outside the burrow between three and twelve days after definitive molts (this agrees with Baerg ’s field obser- vations). Most natural bun'ows, however, did not appear to have the necessary space for such activity. Baerg observed individual males constructing as many as 1 7 sperm webs in the course of six weeks and others as few as one in their mature life. From eight males I reared to maturity the number of webs con- structed per individual varied from one to four. One male produced two webs within four days while a second produced four webs with- in 19 days, both without exposure to females. Several field collected males produced one or more webs in captivity while others failed to spin a single web. One male collected 28 July 1992 spun two webs without exposure to a female, one shortly after confinement and a second almost a year later in June 1993 (this specimen, incidentally, seemed to be an ex- ception to the general rule of longevity among males of most Aphonopelma species; even captive males rarely last through the year in which they matured). Initial sperm webs of three captive-reared A. mojave males were produced within 10-21 days after definitive molts; only one produced a second web (date unknown). One field-col- lected male produced a maximum of three webs, each spun within two days after suc- cessive matings; a second male produced two webs without introduction of a female while a third male produced one web two days after mating but failed to produce another web after PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 171 mating with second female. Other males spin- ning only one web were never exposed to fe- males in the laboratory. Seasonal mating activity. — Within tem- perate region Aphonopelma there are two ba- sically distinct breeding seasons: one com- mencing in summer and the other in fall. Males of the summer breeding species gen- erally search for females between mid- July and early September. Males of most fall breeding species are first seen in middle or late September but are infrequently found af- ter mid-November. However, in two fall breeding species (A. paloma and an unde- scribed species from SE Arizona) the onset of male activity is delayed until late October or early November; most activity has ceased in these species by late November and by mid- December (late December in undescribed spe- cies) males are rarely seen. I have noticed that within most California and Arizona Aphono- pelma, irrespective of breeding season, strag- glers are occasionally found long after other conspecifics have perished. Early males (Baerg 1958), on the other hand, seem to oc- cur only in the fall breeding species but are found in significantly reduced numbers from those occurring during the fall months. Two of the three Mojave species, A. iodium and A. mojave, are fall breeders; early males are known to occur only in A. iodium. The third species, A. Joshua, is strictly a summer breed- er. A. mojave males were seen searching for females from early October until nearly the end of November. The earliest collection rec- ord for a male is 5 October (1973, W. Iceno- gle) and the latest, 24 November (1991). All observed mating activity was diurnal. Males were found between 0800 h and late afternoon shortly before sunset; a single male was col- lected after sunset on a warm, humid evening (25 °C), On cooler days (when early morning and late afternoon temperatures were below 15 °C) males became active later in the morn- ing and took shelter earlier in the evening. A. iodium males (fall males only) were col- lected between 16 September (BDM) and 29 November (southern Nevada, by W. Icenogle). Breeding activity in this species is believed to be primarily diurnal although some nocturnal mating may also occur; the majority of males were collected between 0900-1600 h, a few as late as 2230 h on warm evenings (over 20 °C). Summer males, believed to be nocturnal breeders, were collected in August (of three different years in JTNM) and only after dark between 2200-0100 h. Baerg believed that early males were survivors from the previous year that had somehow managed to overwin- ter and that their success in mating at this time was doubtful. All summer males collected in JTNM were in excellent condition and ap- peared to have molted quite recently. Since Baerg observed that mature female and im- mature Arkansas tarantulas begin their molt- ing cycles toward the end of July and I have recorded molting periods of captive A. iodium from mid-June to mid-September, it seems quite possible that these summer males had not overwintered but had, instead, recently matured. Since many females have molted and become active by August it seems likely that early males are successful in their breeding attempts. Observations of two captive males (molting 2 August and 8 September, respec- tively) suggest that differences in temporal spacing between maturing molts and burrow abandonment and in the timing of the molt itself, in combination, may account for both the presence of summer males and the dura- tion of the primary breeding interval. Follow- ing its September molt the latter male re- mained in its burrow for six weeks, emerging only for relatively short periods of time to eat, drink, and, eventually, construct its first sperm web. Its behavior in conjunction with its mat- uration date may be typical of circumstances leading to the emergence of the more preva- lent fall males. On the other hand, the former male became very active within two weeks of its early August molt, rarely returning to its burrow. Its early maturation and hastened rest- less behavior may be indicative of events leading to the presence of the much less com- mon summer males. In captive males pro- longed lingering between definitive molts and burrow abandonment is the more commonly observed behavior. A. Joshua is the only Mojave Desert species known to be strictly a summer breeder. Of the males that I have collected the earliest was taken 20 July (1989), the latest 6 September (1992). However, E.L. Sleeper and S.L. Jen- kins collected one male 9 September (1966) and a second 23 September (1967), both from pit-traps in JTNM. All observed mating activ- ity was nocturnal. Males were collected be- 172 THE JOURNAL OF ARACHNOLOGY tween 2100-0245 h (only two after 0100 h) and were generally seen in greater numbers on warm, humid evenings. Mating behavior. — Most mating behaviors such as male courtship, female response, and post contact performance were corresponding- ly indistinguishable between the desert species and from the respective behaviors of A. pa~ loma (Prentice 1992). Duration of copulatory contact under natural conditions varied within all species although the respective ranges were very similar; the maximum time value of sustained contact for each species was ob- served under laboratory conditions. Male ex- ploration of contacted burrows varied slightly with the entrance type; males generally locat- ed non-turreted burrow entrances more quick- ly than those atop turrets. Males stridulated periodically during frequent pauses in their search for female burrows. In their initial inspection of female burrows typical A. mojave males slowly circled the outer turret wall, systematically pausing with palps in direct substrate contact as if chemi- cally assessing female receptivity. Whether turrets function as pheromone beacons is not known although males sometimes left turreted burrow after brief investigation without initi- ating courtship. Males continuing their in- spections usually proceeded toward the top of the turrets, frequently arresting their forward progress to alternately stridulate (character- ized by bobbing up and down) and forcefully tap the turret walls with both front legs and palps simultaneously. Tapping and stridulation were executed with more regularity once fe- male entrances were located, with one to sev- eral spaced taps alternating between stridula- tory pulses. If there was no immediate female response, males frequently extended their forelegs into the turret or crawled partially or completely inside while continuing to tap and to stridulate if the hind two pairs of legs were free of the entrance. At various stages of courtship, receptive females generally emerged. A drumming response (Prentice 1992) by the female before emergence was observed in several instances when females could be seen in their burrows and in captive situations after male courtship had been ini- tiated. Female drumming (with both pairs of forelegs) was always observed to follow male stridulation but not leg tapping. Once contact between a pair was made, the female com- monly rushed the male pushing him back- wards a few centimeters until he secured her fangs. In one instance, a Red Mountain fe- male, courted by a male having his front legs inside the burrow, exploded from the entrance backing the male almost instantaneously to a distance of 20 cm; after much leg grappling the male finally managed to secure the fe- male’s fangs. Copulatory contact in several field matings was sustained for slightly less than one minute to just under three minutes. In pairings of A. Joshua differences in lab- oratory (when females were in burrows) and field behavior could not be detected except in duration of copulatory contact. Contact be- tween several laboratory pairs was sustained for 1-10 minutes. For field observation males were released at night by female burrows un- der artificial, dim light conditions which made accurate observation difficult. Contact under these conditions was sustained for l-lVi min- utes in two pairings. In all observations un- coupling proceeded with the male releasing one of the female’s fangs while simultaneous- ly pulling away to position himself for rapid departure (true of all Mojave Aphonopelma). All other associated behaviors appeared to be identical to those of A. mojave when the court- ed female was in a turreted burrow. Male stridulation was audible in the laboratory when background noises were at a minimum and was nearly comparable in amplitude to that generated by the much larger A. reversum Chamberlin; I have not heard male stridula- tion in the field. Under natural conditions, A. iodium males usually initiated courtship more rapidly after detecting female silk than males of either A. Joshua or A. mojave. At times, females emerged during the initial stages of courtship before the male had physically located the burrow entrance, generally a rapid process for these males. Duration of copulatory contact was maintained for as little as 30 sec and for as long as 3 min; under laboratory conditions contact was sustained for up to 6 min. Fe- males pursued males after uncoupling much more often than did females of either A. Josh- ua or A. mojave although pursuit, in general, was relatively rare. In no field observation (all species) was a male caught by a female; under laboratory conditions, males with limited run- ning space were occasionally caught and killed by females. PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 173 Egg sacs, fecundity, and spiderlings. — Data for A. Joshua were gathered from two females producing egg sacs in the laboratory and from one female guarding an egg sac when collected. One female (carapace length, 6.5), collected 19 April (1989), excavated and promptly plugged a new burrow in mid-May. A cocoon was produced in the burrow be- tween 23-25 June which, when finished, was a wrinkled, spherical mass approximately 12 mm in diameter. By 26 July the egg sac was very swollen and smooth, suggesting that the eggs had hatched. Between the end of July and 12 August darkened forms appeared in- side of the cocoon; third instar (fourth post- embryonic stage; third stage free of chorion; first mobile stage - Galiano, 1969, 1973) de- velop most leg spines and urticating hair patches beneath the semi-transparent integu- ment, molting to fourth instar within the egg sac (pers, obs.; also consult Galiano 1973). On 19 August the first of the young appeared on the outside of the egg case beside a small exit hole. On 21 August approximately 40 young were counted, a few of which were observed outside of the burrow less than 1 cm from the entrance. By 25 August approximately half of the spiderlings were roaming about in the ter- rarium, periodically returning to the burrow to take refuge. More than three-quarters of the young had permanently dispersed by 30 Au- gust. A total of 51 young was counted, all of which had successfully escaped the egg sac. A second female (allotype), collected 21 Oc- tober (1989) from a plugged burrow, produced an egg sac between 22-27 April (1990) in the burrow she had excavated shortly after con- finement; she consumed the eggs two weeks later. Neither of the preceding females was ob- served without at least one of its fore-append- ages in contact with the egg sac (also Minch 1979). A third female (carapace length, 7.60, following its 1993 molt) was guarding an egg sac when excavated from its burrow 30 July (1992). There was a small emergence hole in the cocoon but only two spiderlings were ob- served on the visible surface. Forty-one 4th instar, three 3rd instar spiderlings that failed to molt successfully, and four desiccated eggs were counted when the cocoon was opened. Due to the breached condition of the egg sac, the actual number of viable young may have been higher. Fourth instar spiderlings of the female that produced a cocoon in the labora- tory began to appear outside the egg sac ap- proximately seven weeks after eggs were de- posited. If it may be assumed in this case that field conditions roughly paralleled laboratory conditions, the female producing a cocoon in the wild would have deposited her eggs during early June, seven weeks prior to the appear- ance of the first young. Baerg (1958) observed that females of Dugesiella hentzi deposited eggs in late June or early July with an incu- bation time of 45 and 65 days under labora- tory conditions; Gertsch (1979) roughly esti- mated an incubation time of 6-7 weeks for TocaT tarantulas in general. Dissection of an A. mojave female (cara- pace length, approximately 8.5) revealed a complement of 186 developed eggs and 150- 200 small developing eggs. Three females produced egg sacs in the laboratory but can- nibalized them before they could be removed; all were produce between the last week in June and the second week of July. Of three females that were excavated from their bur- rows 7 October (1989) near Searchlight, Ne- vada, two were accompanied by what ap- peared to be fourth instar young, one with 18 (not preserved), the second with 15 (carapace lengths, 1.7-1. 8). The third female had a com- plement of 20 young, 5 with carapace lengths of 2.5-2. 8, 14 with lengths of 3. 0-3. 3, and one with a carapace length of 4.1. Data from laboratory reared spiderlings suggest that the young of this female were at least second year spiderlings, or third year in case of the largest young, that were still tolerated or even guard- ed by the female. Carapace lengths of reared spiderlings of three species did not reach 2.5 until the second year for the two larger spe- cies, A. behlei and A. iodium, and until the third year for A. Joshua (closely approximat- ing A. mojave in size). Only fecundity data were obtained for A. iodium. The larger of two females (carapace length, 15.8), collected 7 October (1989) near Searchlight, Nevada, was dissected and con- tained an estimated 800-1000 developed and developing eggs. The smaller female (cara- pace length, 11.5) was collected while guard- ing her brood, many of which were on or near the cocoon. My initial estimation of spiderling numbers was approximately twice that of the 65 eventually retrieved; a substantial portion may have escaped into subterranean crevices when the cocoon was removed. Another factor 174 THE JOURNAL OF ARACHNOLOGY that may have lowered the natural count was the possible dispersal of young before the bur- row was disturbed. Nevertheless, data suggest that larger females are capable of producing more offspring. Prey capture and prey. — Tarantulas, as in other poor sighted hunting/ambushing spiders, detect their prey through substrate vibrations generated by movement. Capture methods of all Mojave Aphonopelma were, typically, in- distinguishable. Once vibratory information was received, Aphonopelma usually turned to- ward the direction of the source. If the prey was relatively near and produced vibrations of the appropriate magnitude, it was quickly rushed, scooped toward the spider's unfolding fangs with fore-legs (or with larger prey with all legs), and impaled almost simultaneously. Once the quarry was secured, Aphonopelma typically extended their legs, raising the some- times struggling victim well above the sub- strate, thus minimizing its chances of escape by using the ground surface for leverage. Most spiders then immediately returned to their burrows, inside of which the meals were consumed. If vibratory stimuli were further away, Aphonopelma moved toward the source through a series of discreet advances, reas- sessing direction, proximity, and magnitude of the vibration with each movement of the prey; the attack sequence was the same once initi- ated. Prey was pursued by A. Joshua and A. mojave for distances up to 20 cm (rarely, up to 30 cm) and by A. iodium for up to 50 cm (rarely, up to greater distances) from the bur- row entrance. Both A. mojave and A. iodium were lured out of their burrows (by using a twig or blade of grass to imitate prey move- ment) when nighttime temperatures were as low as 7 °C (2 °C in A. palomu), suggesting that threshold temperatures for feeding are lower than those required for initiation of male wandering and courtship behavior. Sclerotized remains of beetles (primarily in the family Tenebrionidae), small-to-medium size scorpions, spiders of other families, and orthopterans were found in both burrow chambers and excavations of A. iodium; bee- tles comprised the bulk of the recognizable remains. Prey of both A. Joshua and A. mojave included various species of beetles (primarily tenebrionids), harvester ants (Myrmicinae), small orthopterans, and, occasionally, small scorpions. Soft bodied insects, centipedes, and small lizards were consumed by captive Aphonopelma. Evidence of neither congeneric nor conspecific cannibalism was detected among any of the sclerotized food remnants examined although S. Kutcher (16 October 1990) collected a male A. iodium clutching a male A. mojave in its fangs during the species common breeding season. In the laboratory, captive spiderlings past dispersal age occa- sionally cannibalized each other under over- crowded conditions; males with insufficient space to escape were sometimes killed by fe- males after mating and, presumably, would have been consumed if not removed. Isolating mechanisms in sympatric spe- cies.— A widely accepted paradigm is that, for the maintenance of closely related species in sympatry, some form of isolating mechanism must be in place. Because of their adaptive value premating mechanisms are believed to be of greater evolutionary significance than postmating mechanisms. Since the morphol- ogies of both the male palpal bulb and the female spermathecae in Aphonopelma species (those in which male embolus is slender), doubtfully, preclude copulatory success, sig- nificant size differences in concurrently breed- ing species would reduce the likelihood of in- terbreeding attempts both because of physical constraints and differences in the magnitude of vibrational stimuli produced, the latter pos- sibly eliciting a feeding response in the larger species and a flight response in the smaller species. Males may distinguish conspecific fe- male burrows by entrance diameter although chemical cues, undoubtedly, play a more im- portant role in species recognition. One pos- sible mechanism among species of subequal size is a behavioral skew in breeding seasons; with this energy conservative mechanism in place, such species could coexist with mini- mal contact. Field observations suggested the presence of such mechanisms among temperate North American sympatric Aphonopelma. Species of similar size had distinct, non-coinciding breeding seasons as well as morphological dissimilarities in metatarsal scopulation and/ or coloration or in condition of tarsus IV scop- ula (entire or partially divided). Concurrently breeding species were found to be signifi- cantly different in size and generally distinc- tive in carapace and leg (in females) colora- tion and in degree of metatarsal scopulation. PRENTICE— THERAPHOSIDAE OF THE MOJAVE DESERT 175 The following examples illustrate the breeding season/size related mechanisms and associat- ed morphological differences in known sym- patric species (no species other than those re- ferred to have been found in the stated areas of sympatry): (1) A. Joshua and western A. mojave are naiTowly sympatric and complete- ly overlap in size (also sympatric with A. io- dium); the former is a summer breeder, the latter a fall breeder; partial division of tarsus IV scopula, limited distribution of red-orange dorsoabdominal setae, and swollen third fe- mur of males are the obvious character dif- ferences that distinguish A. Joshua. (2) A. io- dium and A. mojave are both fall breeders (also sympatric with A. Joshua near Yucca Valley); the former are usually much larger than the latter (the smallest A. iodium have been found in southern Utah but are substan- tially larger than the largest A. mojave in that region); extent of metatarsal scopulation and carapace coloration are obvious dissimilari- ties. (3) Two species, very similar in size, in- habit the coastal and inland regions south of the southern California transverse mountain ranges; one is a fall breeding 'eutylenum type’ (preliminary data suggest that Chamberlin names for "eutylenum type’ species described from these areas are synonyms), with typical coloration and extensive metatarsal scopula- tion; the other is a summer breeder (A. re- versum Chamberlin) is solid black (unless fad- ed but is still unicolorous), and has limited metatarsus III and IV scopulation. (4) In Ar- izona A. chalcodes Chamberlin and A. behlei Chamberlin are narrowly sympatric in several regions at elevations approaching 1 800 m (fe- males overlap in size while A. chalcodes males are larger); A. behlei (mountain top or high elevation species) is solid black in color, has limited metatarsus IV scopula, and breeds in fall while A. chalcodes (primarily a desert dweller) has accentuated 'eutylenum type’ col- oration, more extensive metatarsus IV scopu- la, and breeds in summer. The following examples illustrate the re- sponses of Mojave Desert tarantula pairs in situations where both supposed conspecific and interspecific pairings were made: (1) In laboratory breeding experiments two A. Josh- ua males were introduced to A. mojave fe- males, and two males of the latter species were paired with females of the former spe- cies. Pairings were made in both August and October to even out seasonal bias. Males of both species began courtship displays and fe- males responded by typical drumming in three instances, two of which were by A. Joshua fe- males. In all four pairings both genders quick- ly withdrew when contact was made by mov- ing rapidly in opposite directions. That post-contact mating attempts were never ini- tiated suggests that each species may have unique contact pheromones that function in species recognition and as a secondary isolat- ing mechanism in areas of sympatry. (2) East- ern and western A. mojave males readily courted females of the differing race both in the laboratory and under natural conditions. Once females indicated receptivity, copulatory contact was made and sustained in all pairings for varying periods of time. (3) In similar ex- periments with 'eutylenum types’ from vari- ous regions of the Mojave Desert courtship and response behaviors were succeeded by sustained contact during which copulation oc- curred; the following males and females were paired: two male from the Beaver Dam Moun- tains with females from Red Mountain (west Mojave) and the Providence Mountains (east Mojave), respectively (the latter pairing was in mid-October at the female’s natural bur- row); a male from Searchlight, Nevada with a Red Mountain female; a male from Joshua Tree National Monument with a female from Lucerne Valley, California. When paired with A. mojave, female A. iodium exhibited only predatory behavior; similar behavior was seen in male A. iodium although they infrequently ignored females; A. mojave exhibited only a flight response. Expected responses to interspecific pairings between sympatric 'eutylenum type’ species and between species similar to A. mojave would be on the order of those seen in pair- ings of A. Joshua with A. mojave. Instead, the above pairs freely mated, supporting both the synonymy of A. iodium and the conspecificity of eastern and western A. mojave. ACKNOWLEDGMENTS I would like to thank the following people for the loan of type material: Dr. Norman I. Platnick at the American Museum of Natural History, New York, Dr. Jonathan Coddington at the National Museum of Natural History (Smithsonian), Washington, DC, Dr. Jurgen Gruber at the Naturhistorisches Museum 176 THE JOURNAL OF ARACHNOLOGY Wien, Austria, and Mr. Paul D. Hillyard at The Natural History Museum, London. For the loan of specimens from personal collec- tions I would also like to thank Dr. David Bix- ler, Wendell Icenogle, and M.E. Thompson. Special thanks are extended to Dr. Norman I. Platnick, Dr. Fred Coyle, Dr. Charles Gris- wold, and Mr. Wendell Icenogle for their valu- able comments and advise in the preparation of this manuscript. LITERATURE CITED Ausserer, A. 1871. Beitrage zu kenntnis der Ar- achnidien-Familie der Territelariae Thorell. (My- galidae Auton). Verhandl. K. K. Zool.-Bot. Ge- sell., Wien, 21:117-224. Ausserer, A. 1875. Beitrage zu kenntnis der Ar- achnidien-Familie der Territelariae Thorell (My- galidae Autor.). Verhandl. K. K. Zool.-Bot. Ge- selL, Wien, 25:125-206. Baerg, WJ. 1958. The Tarantula. Univ. Kansas Press, Lawrence. 88 pp. Bonnet, R 1955-1959. Bibliographia Aranearum. Toulouse, vol. 2, pts. 1-5: 1-5058. Brignoli, P.M. 1983. A Catalogue of the Araneae Described Between 1940 and 1981. Manchester Univ. Press. 755 pp. Chamberlin, R.V. & W. Ivie. 1939. New tarantulas from the southwestern states. Bull. Univ. Utah, 29:1-17. Chamberlin, R.V. 1940. New American tarantulas of the family Aviculariidae. Bull. Univ. Utah, 30,: 1-39. Cooke, J.A.L., VD. Roth & F.A. Miller. 1972. The urticating hairs of theraphosid spiders. American Mus. Nov., 2498:1-43. Coyle, EA. 1971. Systematics and natural history of the mygalomorph spider genus Antrodiaetus and related genera (Araneae: Antrodiaetidae). Bull. Mus. Comp. Zool., 141:269-402. Coyle, F.A. 1988. A revision of the American fun- nelweb mygalomorph genus Euagrus. Bull. Mus. Comp. ZooL, 187:00-00. Coyle, F.A. & TE. Miggs. 1989. Two new species of kleptoparasitic Mysmenopis (Araneae, Mys- menidae) from Jamaica. J. Arachnol., 17:59-70. Galiano, M.E. 1969. El desarrollo postembrionario larval de Grammostola pulchripes (Simon, 1891) (Araneae, Theraphosidae). Physis, 29:73-90. Galiano, M.E. 1973. El desarrollo postembrionario larval en Theraphosidae (Araneae). Physis, 32: 37-46. Gertsch, WJ. 1961. The spider genus Lutica. Senckenb. Biol., 42:365-374. Gertsch, WJ. 1979. American Spiders. 2nd Ed., Van Nostrand Reinhold Co., New York, 274 pp. ICZN. 1991 (June). Opinion 1637. Aphonopelma Pocock 1901 (Arachnida, Araneae): gives pre- cedence over Rhechostica Simon 1892. Bull. Zool. Nomen., 48. Marx, G. 1988. Proc. Ent. Soc. Washington, 1:116. Minch, E.W 1979. Annual activity patterns in the tarantula, Aphonopelma chalcodes Chamberlin (Araneae: Theraphosidae). Nov. Arthrop., A Journal of Arthropod Nat. Hist., vol. 1. Perez-Miles, F. 1989. Variacion relativa de carac- teres somaticos y genitales en Grammostola mol- licoma (Araneae, Theraphosidae). J. Arachnol., 17:263-274. Perez-Miles, F. 1994. Tarsal scopula division in the Theraphosinae (Araneae, Theraphosidae): its sys- tematic significance. J. Arachnol., 22:46-53. Petrunkevitch, A. 1911. A synonymic index— cata- logue of spiders of North, Central, and South America. Bull. American Mus. Nat. Hist., 29:1- 791. Petrunkevitch, A. 1939. The status of the genus Eurypelma (Order Araneae, Family Therophosi- dae). Ann. Mag. Nat. Hist., ser. 11; 4:561-568. Petrunkevitch, A. & collaborators. 1939. Catalogue of American spiders. Part 1. Trans. Connecticut Acad. Arts Sci., 33:133-338. Platnick, N. I. 1989. Advances in Spider Taxonomy 1981-1987. Manchester Univ. Press. 673 pp. Platnick, N. I. 1993. Advances in Spider Taxonomy 1988-1991. NYES and AMNH, 846 pp. Pocock, R.J. 1901. Some old and new genera of S. American Aviculariidae. Ann. Mag. Nat. Hist., ser. 7; 8:540-555. Prentice, TR. 1992. A new species of North Amer- ican tarantula, Aphonopelma paloma (Araneae, Mygalomorphae, Theraphosidae). J. Arachnol., 20:189-199. Raven, R.J. 1985. The spider infraorder Mygalo- moiphae (Araneae): cladistics and systematics. Bull. American Mus. Nat. Hist., 182:1-180. Roewer, C.E 1942. Katalog der Araneae von 1758 bis 1940. Bremen, 1:1-1040. Simon, E. 1890. Liste des especes de la famille des Aviculariides qui habitent L’Amerique du nord. Act. Soc. Linn. Bord., Vol. XLIV, Tome IV:307-326. Smith, A.M. 1994. Tarantula Spiders, Tarantulas of the U.S.A. and Mexico. Fitzgerald Publ., Lon- don, 196 pp. Strand, E. 1907. Aviculariidae und Atypidae des Kgl. Naruralienkabinetts in Stuttgart. Jahresh. Ver. Naturk. Wurttemberg, 63:1-100. Manuscript received 15 June 1995, revised 27 Sep- tember 1996. 1997. The Journal of Arachnology 25:177-181 CALLOBIUS GUACHAMA (ARANEAE, AMAUROBIIDAE): HABITAT, DISTRIBUTION AND DESCRIPTION OF THE FEMALE Richard S. Vetter and Thomas R. Prentice: Department of Entomology, University of California-Riverside; Riverside, California 92521 USA ABSTRACT. On the basis of one male specimen, Callobius guachama Leech 1972 was first established during the familial revision of the Amaurobiidae. We have collected additional specimens of this spider and, herein, provide a description of the female as well as notes regarding the habitat and distribution of this large, montane spider. In the familial revision of the Amaurobiidae (Leech 1972), Callobius guachama was named on the basis of a single mature male spider collected inside a domicile near the foothills of the San Bernardino mountains. We have recently collected additional specimens in natural environments or have had speci- mens sent to the University of California Riv- erside Department of Entomology for identi- fication. In addition to providing habitat and collection information for the species, we de- scribe the female of Callobius guachama for the first time. Leech (1972) states that the internal geni- talia for all amaurobiid genera except Tita- noeca Thorell 1870 are not useful for taxo- nomic differentiation at the species level. Therefore, in remaining consistent with the fa- milial revision, we configure here the ventral and posterior views of the C guachama epig- ynum. Also, only the diagnostic characteris- tics of the male palp (i.e., tibia, median apoph- ysis) were illustrated in the familial revision. As a record of completeness for this species, we include conventional ventral and lateral views of the entire male palp of C. guachama. METHODS All preserved specimens were examined under alcohol and measured with a Wild 5 A microscope fitted with an ocular micrometer; all measurements are in millimeters. If the ab- domen of an alcohol specimen did not appear shriveled or was not damaged in the collection process, body length measurements were taken. Several spiders were collected as im- matures, as reflected in the collection data, but were maintained in the laboratory and exam- ined as preserved mature specimens. Physical description of the female is presented from live specimens as well as preserved material. The physical characteristics for the males ex- amined here follow that of Leech (1972) for the holotype. The acronyms used in this paper are as follows: AMNH-American Museum of Natural History, N. Platnick; BRH-B.R. He- bert (pers. collection); CAS-California Acad- emy of Science, C. Griswold; DEB-D.E. Bix- ler (pers. collection); MCZ-Museum of Comparative Zoology, H. Levi; RSV-R.S. Vetter (pers. collection); TRP-T.R. Prentice (pers. collection); UCR-Entomology Muse- um, University of California-Riverside; WRI- W.R. Icenogle (pers. collection). Callobius guachama Leech Figures 1-6 Callobius guachama Leech 1972: 53, figs. 84a, b, 6 . Male holotype from San Bernardino [San Ber- nardino County], California, in AMNH, exam- ined. Diagnosis. — Callobius guachama can be separated from other species of Callobius Chamberlin 1947 (except C. nevadensis (Simon 1884) and C. severus (Simon 1884) by its larger size and from all species by gen- italic differences. Mature specimens of C. guachama are consistently large whereas oth- er Callobius species (e.g., nevadensis, sever- us, pictus (Simon) 1884 and C. arizonicus (Chamberlin & Ivie 1947) may only occasion- 177 178 THE JOURNAL OF ARACHNOLOGY ally attain this size; most are medium-sized (10-13 mm) spiders (Leech 1972). The males of C guachama have the largest median apophysis (>1 mm) of any Callobius currently known, whereas median apophyses in both C nevadensis and C. severus are 0.9. Additionally, the median apophysis in C. gua- chama has two distinct, subequal, well-round- ed notches, whereas in C. nevadensis the an- terior notch is much smaller than the posterior and in C. severus notches are not rounded and are poorly defined with an indistinct cusp when evident. The female of C. guachama can be distin- guished from all other Callobius species ex- cept C. severus by having a diminutive pos- terior lobe and from C. severus by having ectal margins of lateral lobes (posterior view) that are very robust and not broadly excavat- ed. Description. — Male: Overall length, 1 1- 15, cephalothorax, 6. 6-8. 2 length, 4. 4-5. 7 width (at Leg III). Length of median apoph- ysis, 1.00-1.22. In preserved specimens, ab- domen varies in coloration from gray-to- brown and cephalothorax is sometimes uni- formly orange, lacking the cephalic darkening noticed in the holotype and some of our spec- imens. The male palp is shown in Figs. 1, 2. Considering the diagnostic characteristics of palpal tibia and median apophysis, the male structures were consistent in their appearance with minor variation in size. The median apophysis consistently had two well-rounded notches of equal depth, the width of the an- terior notch being slightly smaller than the posterior and a pointed cusp rising up to sep- arate them. Female: Overall length, 15-20, cephalotho- rax, 7.0-8.4 length, 4. 7-6. 2 width. Epigynum, width, 1.4-1.74, measured across epigynum at the point where anterior margins of the lateral lobe intersect the transverse ectal margins. Color of legs and carapace usually light to dark chestnut, less often yellow-brown, ce- phalic region darker especially in subadults and adults; chelicerae dark usually appearing black; maxillae and labium also dark; tarsus of palp darker than proximal segments; ab- domen usually dark grey with 2-3 pairs of faint, sometimes indistinguishable anterodor- sal light orange-tan markings, most anterior pair usually appearing as longitudinal bands extending posteriad from anterior margin, 2nd and 3rd pair usually subcircular corresponding to the dimples of muscle-impressions; venter dark, similar to carapace except light yellow- ish or tan regions of book lungs. The epigyna (ventral view) with round-to- ovoid median lobe, lateral lobes as long as wide or slightly longer than wide, ectal lobes lacking and posterior lobe diminutive (poste- rior view), round-to-pentagonal in shape, width approximately 14 width of lateral lobe (Figs. 3, 4). Epigynum highly variable, asym- metrical in about half of specimens examined (total = 22, 6 slightly asymmetric, 6 very asymmetric). The posteriad protuberances on the lateral lobes contiguous with the posterior marginal line of the lateral lobe or protruded noticeably beyond. However, consistencies were noted in the diminutive posterior lobe and the robust nature of the lateral lobes in posterior view (Fig. 4). Outlines of epigyna are presented in Fig. 5 to show both variation and asymmetry. Material examined. — Holotype male (AMNH), 25 d 22 $26 imm. CALIFORNIA: Kern County, Tehachapi Mts., Paradise Valley (elev. 5000 ft.), 18 May 1960, 5 June 1960, in house, 2c3, W. Icenogle (WRI). Los Angeles County, San Gabriel Mts. (elev. 4800 ft.), Glendora Ridge, 4 August 1994, under road culvert, 1 $, T. Prentice (TRP); Soldier Creek (elev. 3740 ft.), 11 August 1994, limm. $, T. Prentice (TRP). Riverside County, San Jacinto Mts., Idyllwild (elev. 5400 ft.), 24 June 1969, 1 6 , H.E. Brown (UCR); late November 1996, in house, 1 $ , C. Hamilton (RSV). San Ber- nardino County, San Bernardino (elev. 1150 ft.), 1 July 1969, in house. Id (holotype), R. Miller (AMNH); San Bernardino Mts., Big Bear Lake (elev. 6750 ft.), 13 May 1994, in house. Id; 4 June 1994, in house, 1$, N. Kohl (RSV); 29 August 1994, 1 $ (RSV); no date. Id (RSV); 30 June 1995, Id (RSV); 26 July 1995, in house in web in ceiling corner, 1$, A. Sayles (RSV); 24 October 1995, in home on staircase 0300 h, 1 $ , J. Reisman (RSV); early June 1996, in home in web, 1 d, J. Castiglioni (RSV); Crestline (elev. 5200 ft.), 29 May 1 996, on bedroom ceiling. Id, K. McKinley (RSV); Fish Creek (elev. 6550 ft.), 8 June 1995, under bark, 3$, T. Prentice (TRP); Forest Falls (elev. 7000 ft.), 17 May 1987, in house. Id (DEB); Lake An'owhead (elev. 5000 ft.), 30 April 1991, in cabin, 1 $, M. Laurich (BRH), 13 June 1995, in house. VETTER & PRENTICE— CALLOR/t/5 GUACHAMA 179 Figures 1-5. — Callobius guachama Leech. 1, Male left palp, ventral view. 2, Left Palp, lateral view. ma = median apophysis. 3, Epigynum, ventral view;., 4, Epigynum, posterior view; 5, Outline of epigyna (ventral view) showing variation and asymmetry. IS,F. Kimble (RSV); Lost Creek (elev. 7400 6050 ft.), June 1994, in house, IS (RSV); 6 ft.), 17 May 1995, under fir stump, 200 ft. June 1995, 2d Ipenult? (RSV); 28 July 1995, from creek, 1$, T. Prentice (TRP); Mountain in house, 1?, S. Swinson (RSV); Santa Ana Home Creek, E. fork (elev. 5000 ft.), 29 River (elev. 5500-6550 ft.), 15 November March 1995, 2dl94imm, 26 April 1995, 2$, 1994, 1 $ Ipenultd; 30 March 1995, T. Prentice (TRP); Running Springs (elev. Id3imm$; 10 May 1995, under bark of fir 180 THE JOURNAL OF ARACHNOLOGY Bakersfield l/i 'VAiuEY San Bernardino Los Angeles Riverside ^ ® SANTA AN> Figure 6. — Geographical distribution of Callobius guachama (map from U.S. Geological Survey, Den- ver, Colorado). tree, 2$, T. Prentice (TRP); Seven Oaks (elev. 5600 ft.), 5 May 1996, under garbage can, un- der bark, 4imm, 17-19 May 1996, under loose bark of fallen pine trees, 2 d 1 9 1 2imm, R. Vetter (RSV); Sugarloaf (elev. 7050 ft.), 27 June 1995, in kitchen sink, 19, 3 July 1995, in bathroom, Id, K. Vargas (RSV); Twin Peaks (elev. 5400 ft.), no date, Id, W, Sears (UCR), 22 April 1996, in bathroom, Ipe- nultd, 23 May 1996, in toy chest. Id, C. Wormald (RSV); June-July 1996, under trash, 3imm, K. Wormald (RSV); 13 July 1996, in house. Id, C. Hinkleman (RSV). We have deposited several specimens of each sex at both AMNH and CAS. Most of the remainder are deposited at UCR or TRP collections. DISCUSSION Callobius guachama is a montane spider found in southern California from 1150-2250 m elevation on at least four mountain ranges (San Jacinto, San Bernardino, San Gabriel and Tehachapi; Fig. 6). The last three of these ranges are contiguous, while the more south- ern San Jacinto mountains are separated from the nearby San Bernardino mountains by a narrow pass of 600 m elevation. In contrast to the other specimens we ob- tained in this study, the holotype male was VETTER & PRENTICE— CALLO^/f75 GUACHAMA 181 described from the densely-populated urban area (Norton Air Force Base, elev. 300 m) near the foothills of the San Bernardino moun- tains. All specimens presented to us by the public have come from the sparsely-populated mountain communities. If C guachama does live in the lowlands, it is surprising to us that more specimens have not been turned in to authorities for identification. It appears pos- sible that the holotype may have been trans- ported from an area of higher elevation to its collection site. Callobius guachama is found in natural rock outcroppings, under bark of dead fir and pine trees, in deep crevices of living cedar, pine and fir or in human-altered environments such as under road culverts and highway un- derpasses. It also is occasionally discovered in domiciles in the mountain communities, at times causing great alarm to the human in- habitats who fear that this large spider is dan- gerous. Most C. guachama males in this study were found in homes (probably searching for mates) and were subsequently destructively captured and brought in for identification, but a few females were also similarly collected. Most of the spiders collected by the lay com- munity were found in the warm months of May-August, with one female being taken in a house in late November; our field-collected spiders were taken from late March to No- vember. The areas in which these spiders have been collected have winter temperatures rou- tinely below 0 °C with extended periods of snow cover (some of these locales are popular winter tourism areas). Callobius guachama has been active when temperatures were as low as 8 °C; one spider was taken at dawn while it crawled on a wall in an unheated campground washroom. Using the Callobius key provided in Leech (1972), all of the males in this study emerge as guachama. Females of C guachama uni- formly can be keyed out to having two ventral (4 total) spines distally located on metatarsi I and II (couplet 25b) and usually 3 or 4 spines proximally on metatarsi I and II (couplet 40a) although some had 1 or 2 metatarsi with as few as 2 and as many as 6 spines. From this point, the posterior lobe of the epigynum is diminutive which would key out to C. severus (couplet 41b; one of two times the female of this species emerges in the key). (We use here the term “diminutive” whereas Leech used the term “vestigial”. As a reviewer correctly pointed out, this latter term denotes an evo- lutionary derivation of a structure once func- tional). One might amend the key to incor- porate the female of C. guachama by adding at this point, “in posterior view, diminutive posterior lobe with robust lateral lobes”. Finally, because of the depauperate Callo- bius fauna in southern California, we investi- gated the possibility that the holotype of Aux- imus pallescens Chamberlin 1919 might be an immature of C. guachama. Auximus palles- cens was named on the basis of an immature female collected in Claremont near the foot- hills of the San Gabriel mountain range (Chamberlin 1919); it was synonomized with C nevadensis by Leech (1972). We have ex- amined this holotype as well as the four known C nevadensis specimens (all mature females; three from AMNH, one from DEB) from the Los Angeles Basin and are satisfied that the holotype is not an immature of C. guachama and, therefore, no taxonomic name change needs to be made. ACKNOWLEDGMENTS Laura Merrill (USDA Forest Service, San Bernardino) merits a massive quantum of thanks for her encouragement, interest, and dogged persistence in acquiring the necessary scientific collection information from the montane residents of San Bernardino County who have been terrorized in their homes by this large spider. We thank N. Platnick (AMNH) and C. Griswold (CAS) for loan of Callobius material and H. Levi (MCZ) for loan of the holotype of Auximus pallescens. W Icenogle, S. Frommer and an anonymous reviewer made valuable comments which im- proved the manuscript. This study was funded in part by Humbug Mountain Engineering Services R&D fund (P-62) and Parke-Davis Research. LITERATURE CITED Chamberlin, R.V. 1919. New California spiders. Pomona College J. Entomol. Zool., 12:1-17. Leech, R. 1972. A revision of the Nearctic Amau- robiidae (Arachnida: Araneida). Mem. Entomol. Soc. Canada, #84, 182 pp. Manuscript received 8 May 1996, revised 10 No- vember 1996. 1997. The Journal of Arachnology 25:182-193 FORAGING VERSATILITY AND THE INFLUENCE OF HOST AVAILABILITY IN ARGYRODES TRIGONUM (ARANEAE, THERIDHDAE) Karen R. Cangialosi: Dept, of Biology, Suite 2001, Keene State College, Keene, New Hampshire 03435-2001 USA ABSTRACT. Argyrodes trigonum (Hentz 1850) can interact with its host as kleptoparasite, host pred- ator, web-stealer, or commensal. This species can also capture insect prey in a web of its own construction. Which foraging strategy an individual A. trigonum exhibits certainly depends on a multitude of environ- mental factors, especially host availability. In this study, field surveys of populations of A. trigonum and its hosts and daily observations of individually marked host webs were made at sites in Ohio and New Hampshire. These observations together with a manipulation of A. trigonum density were performed in order to determine the influence of host species and abundance on the foraging strategy of A. trigonum. A. trigonum utilized Neriene radiata (Walckenaer 1841) to a greater extent than alternative hosts at both web sites even though many other host species were more abundant. The percentage of A. trigonum sharing a web with the host did not change with differing host/A. trigonum ratios; however, as a host/A. trigonum ratio increased, more A. trigonum were found in unoccupied host webs and fewer A. trigonum were found in webs of their own construction. A. trigonum is more likely to share a web with Pityohyphantes costatus (Hentz 1850) and to usurp the webs of Neriene radiata. Overall, A. trigonum behaved predominantly as a host predator; however, kleptoparasitism is more likely in host webs that last longer. Capturing prey in self-constructed or empty host webs is also important to A. trigonum foraging. While some species utilize only one or a narrow set of behaviors to accomplish a cer- tain task such as acquiring food, others exhibit a broad repertoire of behavioral strategies to achieve the same goal. Such behavioral ver- satility in a population could be the result of phenotypic plasticity of individuals respond- ing to a variety of environmental pressures, genetic differences among individuals within a population, or both. Although some species of spiders are known to be quite versatile in their behavior (Jackson & Hallas 1986; Jack- son & Poulsen 1990), very little research has been done to determine the correspondence between alternative behavioral strategies and their possible associated environmental fac- tors. The spider genus Argyrodes Simon 1864 (family Theridiidae) is commonly thought to be comprised of species that forage primarily by invading the webs of other host spiders and kleptoparasitizing their captured prey, or be- having as commensals in host webs. However, some species have also been shown to be predators of their hosts (Exline & Levi 1962; Smith-Trail 1980; Wise 1982; Tanaka 1984; Larcher & Wise 1985; Whitehouse 1986; Su- ter et al. 1989). According to Whitehouse (1986, 1987), those species that appear to be exclusively araneophages (or host predators) are members of the sub-genera Rhomphaea (L. Koch 1872) or Ariamnes (Thorell 1870), which may actually be genera distinct from Argyrodes (although closely related). In fact, many species of Argyrodes are both klepto- parasitic and araneophagic (Whitehouse 1986), and the foraging behavior of few spe- cies has been studied in enough detail to de- termine the full range of their foraging pos- sibilities (but see Vollrath 1979a, b: A. elevatiis Taczanowski 1873; Whitehouse 1986, 1988, 1993: A. antipodiana Cambridge 1880; and Cangialosi 1990, 1991: A. ululans Cambridge 1880). Work by Larcher & Wise (1985) and Cangialosi (unpubl. data) indicates that Argyrodes trigonum (Hentz) utilizes an array of foraging tactics including kleptopar- asitizing prey from a host web, using an oc- cupied or unoccupied host web to capture its own prey, preying on the host spider, and cap- 182 CANGIALOSI— FORAGING BY ARGYRODES 183 turing insect prey in a web of its own con- struction. Elucidating the factors responsible for foraging versatility in A. trigonum should further our understanding of the ways in which the environment may or may not influ- ence behavior. Although many factors are probably influ- ential in determining which foraging strategy is exhibited by an individual A. trigonum, the availability of hosts is presumably one of the most important. The major objectives of this investigation were to determine the diversity of host species utilized by A. trigonum, the relative importance of each of the different foraging modes it exhibits, and how foraging mode is influenced by host species and host abundance. In particular, I hypothesize that 1) changes in host abundance cause shifts in the percentages of A. trigonum exhibiting differ- ent foraging strategies, and 2) A. trigonum uses different foraging tactics when interact- ing with the host species, Neriene radiata Walckenaer (family Linyphiidae) than it does when interacting with the host species. Pit- yohyphantes costatus Hentz (family Linyphi- idae). These two host species were selected for comparison because both are major hosts for A. trigonum, and differences between them in web structure and body size was expected to provide different foraging challenges for A. trigonum. METHODS Study sites. — Two study sites were used for data collection and comparison. One site was the forested portion of Miami Universi- ty's Ecological Research Center in Oxford (Butler County), Ohio. The other was the Greater Goose Pond Forest in Keene (Chesh- ire County), New Hampshire. Both of these sites are temperate deciduous forest although the Keene site has a greater proportion of co- niferous trees, especially white pine and hem- lock, Argyrodes trigonum and its hosts are common in the understory of both forests. Study species.— -A trigonum is common throughout the eastern portions of Canada (Ontario), and the United States from central Wisconsin to eastern Texas, and Maine to Florida (Exline & Levi 1962). The body length of adults ranges from approximately 2- 4 mm. When not in a host web, A. trigonum builds a very small tangle web or hangs from just a few strands of silk. The two host species used in the individual web observations, Neri- ene radiata and Pityohyphantes costatus, are both linyphiids. Neriene radiata builds a dome-shaped sheet web with barrier silk ex- tending above the dome. The spider usually sits just beneath the central area of the dome. Pityohyphantes costatus builds a hammock- shaped triangular sheet which is flatter and longer than that of N. radiata. Barrier silk also extends above the sheet of P. costatus. Pit- yohyphantes costatus builds a retreat which usually consists of dense silk placed in a rolled leaf or under a piece of tree bark at one end of the sheet. The spider is often found within this retreat, or underneath the central part of its sheet web. Ohio site field survey. — To gain some measure of overall host use by A. trigonum in this study site, a 20 X 2 m plot of forest was censused weekly from August-October 1990 for a total of 1 1 weeks. The number of web- building spiders of all species (if easily iden- tified) or family present was recorded. I also recorded the presence of A. trigonum in a web with a host, in a host web alone, or in a web of its own construction. Then, the number of host spiders relative to the number of A. tri- gonum (host/A. trigonum ratio) was calculated for each date. The percentage of A. trigonum observed in each of the three above situations was plotted against host/A. trigonum ratio for all 11 dates and Spearman Correlation Coef- ficients were calculated. New Hampshire site \iostJ Argyrodes ratio manipulation. — In the forests of New Hamp- shire, it is common to find short walls of piled stones (mostly granite) that were used as prop- erty dividers 100 or more years ago. Many understory spiders build webs on these rocks and the vegetation that grows between and around the rocks. I utilized one of these walls as a convenient way to define control and ma- nipulated areas. The wall used was approxi- mately 0.75 m high. Three areas along the wall, each 10 m in length and 1.5 m in width, were marked at the edges with painted tent stakes and randomly designated as control, re- moval or addition. The three areas were sep- arated by approximately 40-50 m of stone wall that was ignored in this study. In order to create a wide range of ho^Xi Argyrodes ra- tios, I manipulated A. trigonum density in two of the three areas. I removed all A. trigonum from the removal area beginning on 13 July 184 THE JOURNAL OF ARACHNOLOGY Table 1. — Argyrodes trigonum utilization of host webs at the New Hampshire and Ohio study sites. “With A. trigonum'' indicates web sharing, (arg = A. trigonum). Data are from the control area of the NH site density manipulation and the Ohio site field survey. New Hampshire Ohio # of # with % occu- # arg in # of # with % occu- # arg in hosts A. tri- pied webs host web hosts A. tri- pied webs host web Host spider alone gonum with arg alone alone gonum with arg alone Linyphiidae Neriene radiata 471 16 3.3 44 277 30 9.8 92 Frontinella pyramitela — — — — 190 5 2.6 13 Pityohyphantes costatus 70 1 1.4 2 34 1 2.8 2 Other Linyphiidae 192 0 0 0 — — — — Theridiidae 782 0 0 0 — — — — Agelenidae 297 1 0.34 0 11 4 26.7 2 Orb Weavers 62 0 0 2 341 0 0 0 1993 and continuing every 1-2 days until 10 October 1993. A. trigonum were added to veg- etation in the center of the addition area, but not directly in host webs, and kept at the level of 10-15 total individuals (checked every 1- 2 days) in this same time period. The A. tri- gonum used for additions were those taken from the removal area as well as some spiders collected approximately two km away from the experimental areas. The control area was left alone. The foraging situation of A. trigonum (shar- ing a web with host, in host web alone, or in self constructed web) was recorded for all in- dividuals within the three areas every 1-2 days. Number of hosts and A. trigonum were also recorded in each of the three areas and the host/A. trigonum ratio was calculated for each observation date. The number of A. tri- gonum in the removal area was not zero be- cause of continuous immigration of A. trigo- num into this area. Observation of their foraging situation was made just before re- moval. As with the Ohio data, the percentage of A. trigonum observed in each of the three above situations was plotted against weekly host/A. trigonum ratio and Spearman Con*e- lation Coefficients were calculated. (Weekly, instead of daily, ratios were used in order to make more direct comparisons with the Ohio data. I used the ratio for the first day of the week that counts were recorded). Because a wider range of host/A. trigonum ratio was ex- hibited in these manipulated areas compared to the Ohio site, two sets of Spearman corre- lations were performed for the New Hamp- shire data: one for host/A. trigonum ratios of less than 6:1 (for comparison with the Ohio data), and another for all ratios. Additionally, the data from the control area was compared to the Ohio site field survey in order to com- pare overall host species utilization between the two sites (Table 1). Observations of individual host webs. — At the New Hampshire site, occupied webs of Neriene radiata and Pityohyphantes costatus were located and the web site and webs were individually marked by placing flagging on vegetation near the web and a twist tie at one edge of the web at its attachment to the veg- etation. No spiders were marked. Observa- tions of groups of 23-25 host-only occupied webs of each species were initiated on 15 June, 19 July, 9 August, and 26 August 1994, making a total of 94 P. costatus and 95 N. radiata webs that were observed. Each web was observed every day until the complete disappearance of the web. The following data were recorded: host was alone in its web, A. trigonum was alone in the host web, the host and A. trigonum were together in the web, the web was empty (no spiders), the web was de- stroyed or gone. If an A. trigonum invaded a host web, emigrated, and then another (or the same) A. trigonum invaded that web later, the host web was considered to be invaded twice. Because several host webs were invaded by A. trigonum more than once, the total number of observations beginning with a host alone in its web was 148 for P. costatus, and 107 for N. radiata. I initially summarized these observations CANGIALOSI— FORAGING BY ARGYRODES by constructing an ethogram of all fates of host webs with respect to the invasion of A. trigonum. Then, frequency of transition (%) from one state to the next was calculated be- tween all states (i.e., host alone, Argyrodes alone, Argyrodes and host together, etc.) for both host species. These frequencies were compared between the two host species using contingency table analysis for the following: the frequency of web sharing and web take- over, the outcome of web sharing (the fre- quency of A. trigonum emigration and host emigration); and the outcome of web takeover (the frequency of host reclaiming the web and A. trigonum emigration). Mean duration of web sharing and web takeover were compared between the two host species using Kruskal- Wallis tests. I also calculated the frequency of empty web invasion by A. trigonum, and compared the persistence (mean duration) of empty webs invaded and not invaded by A. trigonum between host species by log-transforming the non-normally distributed data and then utiliz- ing a 2-way ANOVA. Additionally, mean du- ration of occupied host webs was compared between host species with a Kruskal- Wallis test. RESULTS Host utilization. — Data from the Ohio site field survey and the control area of the New Hampshire site density manipulation were used in Table 1 to compare overall host spe- cies/family utilization between these two sites. A. trigonum uses a variety of hosts; however, a preference for Neriene radiata was seen in both the New Hampshire and Ohio study sites. The percentage of A. trigonum observed shar- ing a web with N. radiata was 2-9 X higher compared to most of the other hosts (Table 1). Also, the number of A. trigonum in empty webs (no host present) was 7-50 X higher in the webs of N. radiata compared to the other host spiders. A. trigonum also made substan- tial use of Frontinella pyramitela (Walckenaer 1841)(family Linyphiidae) at the Ohio site, and Pityohyphantes costatus at both sites (Ta- ble 1). The number of agelenids in the study area at the Ohio site was only 1 1 , but nearly a third of these were observed with an A. tri- gonum individual in their webs. Although there were several hundred agelenids, other linyphiids and theridiids observed at the New 185 Hampshire site, A. trigonum made little or no use of these hosts. Orb weavers were never observed sharing a web with A. trigonum, and only two empty orb webs contained an A. tri- gonum individual (Table 1). Host abundance and foraging mode. — A. trigonum foraging mode was influenced by the relative number of hosts available in some cases in both the Ohio survey and the manip- ulation at the New Hampshire site. Due to the manipulation and the smaller area sizes, the range of host/A. trigonum ratios was much greater in the New Hampshire site (from 1.4: 1 to 67:1 for New Hampshire, and all less than 5:1 in Ohio). There were no significant rela- tionships between host/A. trigonum ratio and any of the three foraging situations at the New Hampshire site when the full range of ratios are included in the analyses. However, when host/A. trigonum ratios of less than 6:1 are considered, some patterns emerge that are similar to the Ohio site data. At the Ohio site, the percentage of A. tri- gonum observed in a web of its own construc- tion decreased significantly with an increase in the host/A. trigonum ratio (Spearman Coeff. R = —0.644, P < 0.05, Fig. la). This same pattern was seen at the New Hampshire site (Spearman Coeff. R = —0.769, P < 0.001, Fig. lb). At the Ohio site, the percentage of A. trigonum observed in host webs alone in- creased significantly with an increase in the host/A. trigonum ratio (Spearman Coeff. R = 0.725, P < 0.01, Fig. Ic). However, this re- lationship was not seen at the New Hampshire site (Spearman Coeff. R = 0.449, P = 0.192, Fig. Id). There was no relationship between the percentage of A. trigonum sharing a web with a host spider and the host/A. trigonum ratio at either site (Ohio: Spearman Coeff. R = —0.198, P > 0.10, Fig. le; New Hamp- shire: Spearman Coeff. R = 0.056, P > 0.5, Fig. If). Host species and foraging mode. — The observations of N. radiata and P. costatus webs at the New Hampshire site revealed sev- eral sequences that took place with respect to the invasion of A. trigonum from the time that a host was first observed occupying its web alone until that web’s demise. These sequenc- es are summarized as a whole in Fig. 2 and subsets of this figure appear in Figs. 3, 4. Overall, a high percentage of host webs were invaded by A. trigonum. Of the total number 186 THE JOURNAL OF ARACHNOLOGY a. b. 100 Self-Constructed Web (New Hampshire) 80 Z UJ o t£ UJ Q. 60 40 e 20 - 0 1 L 1 2 e • —I I I I 3 4 5 6 HOST/ARG RATIO c. cl ‘ Host Web Alone (New Hampshire) h- z UJ o cr UJ Q. 20 - ■ e 0 , 1 , 1 , 1 , 1 , 1 1 2 3 4 5 6 HOST/ARG RATIO luu r 80 60 40 e . 100 r 80 S 60 o QC UJ Q. 40 f. Web Share (Ohio) 3 4 HOST/ARG RATIO Figure 1. — Percentage of Argyrodes trigonum found in unoccupied host webs (host web alone), occupied host webs (web share), or in self-constructed webs, compared to relative host abundance for the Ohio survey (a, c, e) and the New Hampshire density manipulation (b, d, f). ARG = Argyrodes trigonum. of observations beginning with a host alone in its web, 45.9% of P. costatus webs and 54.2% of N. radiata webs were invaded by A. trigonum (Fig. 3). There were 30 webs that were invaded more than once. For P. costatus. 17 webs were invaded twice, three webs were invaded three times, and one web was invaded four times. For N. radiata, seven webs were invaded twice and two webs were invaded three times. CANGIALOSI— FORAGING BY ARGYRODES 187 Figure 2. — Fates of host occupied, Argyrodes trigonum occupied, both occupied, and empty webs for individual webs observed at the New Flampshire site. (This figure can be compared to Figs. 3 & 4 to determine percentage outcome for observations beginning with the rectangular boxes.) N. radiata P. costatus □ Host Emigrates □ Web Share ■ Web Takeover Figure 3. — The percentage of total host-only oc- cupied webs that were either taken over by an Ar- gyrodes trigonum, shared with an Argyrodes tri- gonum, or host emigrated/web destroyed for both Neriene radiata (n — 107) and Pityohyphantes cos- tatus (n = 148) webs (x^ = 30.97, P < 0.0001). N. radiata P. costatus 0 Arg Emigrates □ Both Emigrate ■ Host Emigrates Figure 4. — Web sharing outcome. The percent- age of host and Argyrodes trigonum occupied webs that resulted in Argyrodes trigonum emigration, host emigration, or both emigrating for both Neri- ene radiata {n = 22) and Pityohyphantes costatus (n = 57) webs (x^ = 7.73, P < 0.05). 188 THE JOURNAL OF ARACHNOLOGY 45 NUMBER OF DAYS Figure 5. — Web sharing time. Frequency distribution of time (days) Argyrodes trigonum spent in oc- cupied host webs for both host species. Web sharing vs. web takeover: In order for A. trigonum to eventually usuip a host’s web, there must be some period of co-occupation of the web with that host. Web takeover here refers to an A. trigonum individual observed to be alone in a host web within 24 h of that web having been observed with the host as the sole occupant, while web sharing refers to the host and A, trigonum co-occupying the web for at least 24 h. The invasion of occu- pied host webs by A. trigonum more frequent- ly resulted in web takeover for N. radiata and in web sharing for P. costatus (x^ = 30.97, P < 0.0001, Figs. 2, 3). Web takeover may in- dicate either host predation or web stealing (through forced host emigration). Although webs generally were not observed for more than a few minutes on each day, I did observe direct evidence of host predation on several occasions. There were four observations of N. radiata being fed on by A. trigonum, or a dead N. radiata in the web next to an A. trigonum, and one observation of A. trigonum feeding on P. costatus. All of these webs had a hole tom in the dome or sheet approximately 2-4 cm in diameter. Most of the host webs that were seized by A. trigonum were observed with large holes in the dome or sheet portion of the web. I also observed the host being chased off its web by A. trigonum a total of two times, once for each of the two host spe- cies. The reverse situation was observed once when an A. trigonum individual was chased off the host web by P. costatus. Outcome and duration of web sharing: There was a significantly higher proportion of A. trigonum only emigrating from P. costatus webs, and a significantly higher proportion of both host and A. trigonum emigrating from N. radiata webs after a period of web sharing (x^ = 7.73, P < 0.05, Figs. 2, 4). There was no difference between the two host species in the frequency of the host giving up the web and leaving A. trigonum alone after web sharing occurred. Also, mean duration of web sharing was not significantly different for P. costatus compared to N. radiata (Kruskal- Wallis: x^ — 0.195, P = 0.659, Fig. 5). Whereas web shar- ing never lasted longer than five days for N. radiata, there were two observations of web sharing for P. costatus that continued for a greater period of time than this, one for seven days and one for 10 days (Fig. 5). Outcome and duration of web takeover: CANGIALOSI— FORAGING BY ARGYRODES 189 NUMBER OF DAYS Figure 6. — Time alone in host web. Frequency distribution of time (days) Argyrodes trigonum spent in unoccupied host webs for both host species. Once an A. trigonum had become sole occu- pant of a host web, the likelihood of a host regaining control of that web was small for both host species (less than 10% of webs). The percentage of webs reclaimed by host spi- ders after being usurped by A. trigonum, was not significantly different between P. costatus and N. radiata (x^ = 1.34, P > 0.24). The mean duration of A. trigonum alone in a host web was not significantly different for the two types of host webs (Kruskal- Wallis: X" = 1.17, P = 0.278, Fig. 6). The longest periods of time that an A. trigonum spent alone in a host web were in N. radiata webs. There were three observations of an A. trigonum spending 10, 12, and 45 days alone in a N. radiata web (Fig. 6). Table 2. — ANOVA for mean duration (log-trans- formed) of empty host webs invaded and not in- vaded by Argyrodes trigonum. Source df Sum of squares F-ratio P > F Host species 1 0.621 6.722 0.010 Invasion 1 1.665 18.023 <0.0001 Species X invasion 1 0.030 0.326 0.568 Persistence of occupied and unoccupied webs. — The average time beginning with a host first observed in its web until the host was gone, was significantly greater for P. cos- tatus {P. costatus: mean ± SE — 9.74 ± 1.15 days, n = 70; N. radiata: mean ± SE = 3.74 ± 0.63 days, n = 46; Kruskal- Wallis: X“ “ 13.77, P < 0.001). Host webs that became devoid of a host spider (empty) were some- times invaded by A. trigonum. For P. costatus, 19.4% (18 out of 93) of empty webs were invaded by A. trigonum compared to 9.2% (8 out of 87) of empty N. radiata webs. The per- sistence of a web without its host spider was significantly longer for those webs that were invaded by A. trigonum compared to webs that were never invaded (Table 2). As with occupied webs, empty P. costatus webs lasted significantly longer than N. radiata webs (Ta- ble 2). DISCUSSION Vollrath (1984) classified species of Argy- rodes as either generalists or specialists, where generalists are those species that utilize a wide variety of hosts from different families, and specialists are restricted to one or a few host species. Whitehouse (1988) added that 190 THE JOURNAL OF ARACHNOLOGY generalists use only a few techniques to obtain food, while specialists utilize several tech- niques. However, a host generalist might be expected to need a wider scope of foraging techniques in order to deal with hosts of dif- fering size, defensive ability, and web type. A. trigonum appears to be a generalist in both senses as it uses a variety of hosts and a va- riety of foraging strategies. Although A. trigonum utilized several types of hosts at both the Ohio and New Hampshire sites, its presence in the webs of hosts was not in proportion to the number of those hosts or host webs available. Even as a generalist, A. trigonum shows a preference for certain host types which were mainly linyphiids at these study sites. The eleven agelenid webs at the Ohio site were large (about 30 cm in diameter) and provided an extensive amount of bamer webbing which might explain the extremely high percentage of A. trigonum in their webs. The agelenids, theridiids, and other linyphiids at the New Hampshire site were mostly ju- veniles within the time period of this study, tended to have little banier silk, and built their webs deep in the small spaces between the rocks. These characteristics likely made these hosts less accessible or functional to A. tri- gonum. Overall, the number of A. trigonum at the New Hampshire site was low compared to the Ohio site, especially considering that a greater number of webs at the New Hampshire site were surveyed. In this study, A. tiigonum made virtually no use of orb weavers. At a site in Maryland, A. trigonum is found often in the webs of the orb weaver, Metepeira lab- yrinthea Hentz 1847 (family Araneidae) (Wise 1982; Larcher & Wise 1985). Orb weavers in this genus build a ban-ier web in addition to the orb whereas most others do not. I have never observed M. labyrinthea at the New Hampshire site, and have only seen a few in- dividuals at the Ohio site in the course of sev- eral years. Although A. trigonum exhibits preferences for certain hosts when choices are available, its ability to utilize many different hosts may be largely responsible for its wide geographical distribution. In this study, A. trigonum was observed oc- cupying the web of a host that is no longer present, occupying the web of a host that is present (web sharing), and occupying a self- constructed web. Changes in relative host abundance influence A. trigonum foraging mode to a certain extent by altering the per- centage of A. trigonum found in these three situations. However, determining precisely how individuals shift their mode of foraging is difficult because these situations indicate a complexity of foraging alternatives (Table 3). Also, foraging is certainly influenced by a va- riety of other factors. The abundance data to- gether with the observations of individual host webs reveals more about the relative extent to which A. trigonum behaves as a kleptopara- site, predator, web stealer, or independent for- ager. Several pieces of evidence indicate that A. trigonum behaves as a spider predator or web stealer to a greater extent than a kleptoparasite in these two study areas. More A. trigonum were observed in unoccupied host webs than in occupied host webs at both the Ohio and New Hampshire sites. Also, there was a high- er percentage of A. trigonum in unoccupied host webs than in either of the other two sit- uations (in occupied host webs, in self-con- structed webs) in all three of the areas in the manipulation (addition, removal, control). Be- cause the percentage of total hosts at any one time with A. trigonum in their webs was fairly low (about 1-10%), and the percentage of ob- served webs that eventually were invaded by A. trigonum was high (45.9-54.2%), it seems that relatively few A. trigonum move around quite frequently and eventually invade a large portion of the available webs. This high mo- bility in general is more consistent with a predatory or web-stealing as opposed to a kleptoparasitic lifestyle. The direct observa- tions of host predation support this claim as well. In spite of the largely predatory nature of A. trigonum, the importance of prey klepto- parasitism cannot be ruled out. About 20% of A. trigonum both in the manipulation in New Hampshire and in the Ohio survey were con- sistently observed in an occupied host web (web sharing), regardless of host density. In comparing the two host species, web sharing occuiTed more frequently with P. costatus whereas web takeover was more likely with N. radiata. This is probably related to the fact that occupied P. costatus webs last longer than N. radiata webs and provide a greater amount of banier silk. Host size (Larcher & Wise 1985) and defensive behavior are also important. P. costatus is somewhat larger and CANGIALOSI— FORAGING BY ARGYRODES 191 Table 3. — Three major situations in which an Argyrodes trigonum individual can be found in relation to a host and the modes of foraging and access to foraging sites that these situations indicate. Argyrodes trigonumlhosl Argyrodes trigonum foraging modes and web situation access to foraging sites Occupy host web alone 1) has preyed on the host (host predator) 2) has caused the host to emigrate and is using the host web for insect prey capture (web stealer) 3) has invaded an empty host web and is using it for insect prey capture (web scavenger) Share web with host 1) is taking insect prey unimportant to the host (commensal) 2) is taking insect prey important to the host (kleptoparasite) 3) is in a temporary transition stage to steal host web or prey on the host In self-constructed web 1) is foraging for insect prey (independent forager) 2) is in the process of host web location usually resides in a retreat at the edge of its web under bark or in a rolled leaf. This may explain why a greater percentage of A. trigo- num emigrated from P. costatus webs com- pared to N. radiata webs leaving the host alone again after a period of sharing. Larcher & Wise (1985) also found that the probability and duration of web sharing was different for different host species. Metepeira labyrinthea were less likely to abandon their webs when invaded by A. trigonum compared to N. ra- diata, although A. trigonum did prey on M. labyrinthea at a substantial rate. In general, it seems reasonable to assume that in areas which are dominated by larger host species with long lasting webs and a large amount of barrier or tangle silk, and perhaps reside in retreats (e.g., large agelenids, theridiids such as Achaearanea tepidariorum C.L. Koch 1841), A. trigonum will behave predominately as kleptoparasites or commensals. The significance of capture of their own in- sect prey by Argyrodes (whether by use of a host web or a self-constructed web) as a way of obtaining food, has been minimized or ig- nored by most workers. Vollrath (1984), in his review of kleptobiotic interactions in inverte- brates, even states: “no Argyrodes is known to construct and operate a feeding web”. However, Eberhard (1979) described in detail the use of a self-constructed web by A. atten- uatus in order to capture prey which included not only spiders but a large proportion of in- sects. Although this web differs from the typ- ical theridiid snare, it is used as a substrate for the capture of insect and spider prey and could therefore be considered a capture web. A. an- tipodiana will attack and subdue flies on both its own and the host’s web (Whitehouse 1986). I have observed A. trigonum capture and feed on insects on a self-constructed web both in the laboratory and in the field (unpubl. obs.). Larcher & Wise (1985) showed that A. trigonum captured more than 50% of the in- sects that they introduced into host unoccu- pied webs. In this study, there was always at least 15%, and up to 40%, of the total popu- lation (at both sites) that were found in webs of their own construction. Determining the occurrence of predation versus web-stealing may help clarify the im- portance of self prey capture (capturing its own insect prey) for A. trigonum. Once an A. trigonum had usurped a host’s web, most em- igrated from that web after 1-2 days. Because a web devoid of its host can last about 3-6 days, it appears that predation is usually the goal (whether or not the A. trigonum was suc- cessful). Nonetheless, 29% of A. trigonum in unoccupied N. radiata webs, and 24% of A. trigonum in unoccupied P. costatus webs, stayed for three days or longer with one in- dividual remaining in the same N. radiata web for 45 days. Additionally, since empty webs that are invaded by A. trigonum last longer than those that are not, this implies that either A. trigonum is expending energy in mainte- nance and repair of the web, or staying in 192 THE JOURNAL OF ARACHNOLOGY those webs that happen to last longer. The in- dividual that resided in the same web for 45 days definitely added silk and altered the web considerably so that it was no longer recog- nizable as a N. radiata web. One might view this as stealing the web site rather than just the web. In any case, using the host web for prey capture seems to be an important forag- ing mode for a substantial portion of the pop- ulation. (One note of caution with this con- clusion relates to the fact that the A. trigonum emigration frequency distribution in Fig. 6 follows an exponential decay function. Suter & Sanchez (1991) have presented strong evi- dence that such relationships may indicate a “rolling dice” criterion for decision making, especially if those organisms face an unpre- dictable environment. If this is true here, some individuals may just be randomly waiting lon- ger before moving on to their next predation attempt). Argyrodes almost certainly evolved from web-building ancestors, and their cunent use of self prey capture may still represent a significant amount of their food intake for some species and therefore be more than just a evolutionary vestige. Whitehouse’s (1986) proposed models for the evolution of kleptoparasitism 'm Argyrodes imply that although both araneophagy and kleptoparasitism are present in most species, the foraging behavior of all ancestors and cur- rent species of Argyrodes is dominated by a single strategy. But these data show that there appears to be no consistent dominant foraging mode for A. trigonum, and which strategy it uses depends largely upon the abundance and species of hosts (or prey spiders) that are available. Other environmental factors such as insect availability probably influence A. tri- gonum foraging mode as well and should be investigated in the future. ACKNOWLEDGMENTS I am indebted to several hard-working Keene State College students for their field assistance: Chris Bartlett, Cory Bartlett, Linda Bierweiler, Kelley Endris, Edgar Leighton, and Stella Scott. Ann Rypstra, Sam Marshall and Bob Suter provided many helpful com- ments on the manuscript. This research was supported by faculty development funds and the Division of Science at Keene State Col- lege. LITERATURE CITED Cangialosi, K.R. 1990. Social spider defense against kleptoparasitism. Behav. Ecol. Socio- biol., 27:49-54. Cangialosi, K.R. 1991. Attack strategies of a spider kleptoparasite: effects of prey availability and host colony size. Anim. Behav., 41:639-647. Eberhard, W.G. 1979. Argyrodes attenuatus (Ther- idiidae): A web that is not a snare. Psyche, 86: 407-413. Exline, H. & H.W. Levi. 1962. American spiders of the genus Argyrodes (Araneae, Theridiidae). Bull. Mus. Comp'. Zool., 127:75-204. Jackson, R.R. & S.E.A. Hallas. 1986. Predatory versatility and intraspecific interactions of spar- taeine jumping spiders (Araneae: Salticidae): Brettus adonis, B. cingulatus, Cyrba algerina, and Phaeacius sp. indet. New Zealand J. Zool., 13:491-520. Jackson, R.R. & B.A. Poulsen. 1990. Predatory versatility and intraspecific interactions of Su- punna picta (Araneae: Clubionidae). New Zea- land J. Zool., 17:169-184. Larcher, S.R. & D.H. Wise. 1985. Experimental stud- ies of the interactions between a web-invading spi- der and two host species. J. Aiachnol., 13:43-59. Smith-Trail, D. 1980. Predation by Argyrodes (Theriididae) on solitary and communal spiders. Psyche, 87:349-355. Suter, R.B. & E. Sanchez. 1991. Evolutionary stabil- ity of stochastic decision making in spiders: Results of a simulation. Anim. Behav., 42:921-929. Suter, R.B., C.M. Shane & A.J. Hirscheimer. 1989. Spider vs. spider: Frontinella pyramitela detects Argyrodes trigonum via cuticular chemicals. J. Arachnol., 17:237-240. Tanaka, K. 1984. Rate of predation by a klepto- parasitic spider, Argyrodes fissifrons upon a large host spider, Agelena limbata. J. Arachnol., 12: 363-367. Vollrath, E 1979a. Behavior of the kleptoparasitic spider Argyrodes elevatus (Araneae, Theridi- idae). Anim. Behav., 27:515-21. Vollrath, F. 1979b. Vibrations: Their signal func- tion for a spider kleptoparasite. Science, 205: 1149-1151. Vollrath, F. 1984. Kleptobiotic interactions in in- vertebrates. Pp. 61-94. In Producers and scroungers: Strategies of exploitation and para- sitism. (C.J. Barnard, ed.). Grom Helm, London & Sydney. Whitehouse, M.E.A. 1986. The foraging behav- iours of Argyrodes antipodiana (Theridiidae), a kleptoparasitic spider from New Zealand. New Zealand J. Zool., 13:151-168. Whitehouse, M.E.A. 1987. “Spider eat spider”: The predatory behavior of Rhomphaea sp. from New Zealand. J. Arachnol., 15:355-362. Whitehouse, M.E.A. 1988. Factors influencing CANGIALOSI— FORAGING BY ARGYRODES 193 specificity and choice of host in Argyrodes an- tipodiana (Araneae, Theridiidae). J. Arachnol., 16:349-355. Whitehouse, M.E.A. & R.R. Jackson. 1993. Group structure and time budgets of Argyrodes antipo- diana (Araneae, Theridiidae), a kleptoparasitic spider from New Zealand. New Zeal. J. ZooL, 20:201-206. Wise, D.H. 1982. Predation by a commensal spi- der, Argyrodes trigonum, upon its host: An ex- perimental study. J. Arachnol., 10:11-16. Manuscript received 26 October 1995, revised 10 September 1996. 1997. The Journal of Arachnology 25:194-201 BIONOMICS OF THE SPIDER, CROSSOPRIZA LYONI (ARANEAE, PHOLCIDAE), A PREDATOR OF DENGUE VECTORS IN THAILAND Daniel Strickman': Armed Forces Research Institute of Medical Sciences, Walter Reed Army Institute of Research, Bangkok, Thailand, APO AP 96546 USA Ratana Sithiprasasna: Armed Forces Research Institute of Medical Sciences, Walter Reed Army Institute of Research, Bangkok, Thailand, APO AP 96546 USA Dawn Southard: Dept, of Entomology, University of Maryland, c/o Smithsonian Institution, Dept, of Entomology, NHB-105, Washington, DC 20560 USA ABSTRACT. The pholcid spider, Crossopriza lyoni (Blackvvall 1867) is a common inhabitant of homes in a rural village in Chachoengsao Province, Thailand. Studies on the spider were initiated because its microhabitat closely coincided with that of adult Aedes aegypti (Linnaeus 1762), mosquito vectors of dengue virus. Laboratory observations showed that females deposited eggs 4-6 days after copulation. Females held the egg sac in their mouthparts for 11—13 days, until all spiderlings (mean = 34) had left the sac. Spiderlings did not feed until they had molted, but as soon as feeding commenced they were capable of oveipowering a mosquito many times their own size. Sometimes spiderlings would share a single mosquito or eat a mosquito wrapped by the mother spider. Spiderlings separated from their mother grew more rapidly than those left with the mother and reached maturity in as little as 74 days. The spiders’ principal means of capturing prey was to throw silk with the aid of the hind legs. Spiders used this method to immobilize mosquitoes which were entangled in the standing web or to catch flying mosquitoes. The mosquito was not bitten until the time of feeding, up to six days after capture. Feeding occurred on only 34-48% of the days, and spiders ate about one mosquito per day. Cannibalism was a significant mortality factor, accounting for 67-84% mortality in a cage of spiderlings. An enzyme-linked immunosorbent assay (ELISA) was adapted to test spider tissue for presence of dengue virus. The ELISA was used to show that spiders did not become infected when fed dengue-infected mosquitoes. The results of the study suggested that C. lyoni could form an important component of integrated control of Aedes aegypti mos- quitoes in foci of dengue transmission. Mosquitoes have a tremendous impact on humans almost everywhere, either as signifi- cant sources of imtation or as vectors of se- rious disease. Dengue is the most common vi- ral pathogen transmitted by mosquitoes. The virus, which consists of four distinct sero- types, causes a spectrum of disease ranging from mild fever to fatal shock. Since the late 1970’s, occuiTence of the disease has steadily expanded throughout the tropics and subtrop- ics, to the point that there are millions of cases every year. All confirmed vectors of dengue virus are in the genus Aedes and the most im- ' Current address: 5th Medical Detachment, Unit #15247, APO AP 96205-0020 USA portant vector is Ae. aegypti (Linnaeus 1762) (Gubler 1988). This mosquito thrives in as- sociation with humans, larvae of the species developing in almost any water-filled contain- er (Christophers 1960). In at least some geo- graphical areas, the adult females of Ae. ae- gypti preferentially bite humans indoors (Scott et al. 1993). Spiders can be efficient predators of adult mosquitoes both outdoors and indoors. Stud- ies on spider predation of mosquitoes have ex- amined whether various spider species eat mosquitoes. For example, detailed observa- tions on the rate at which spiders ate mosqui- toes located in large cages in a Polish forest indicated that species of spiders varied in their 194 STRICKMAN ET AL.— CROSSOP RIZA LYONI IN THAILAND 195 appetite for mosquitoes and that the rate of consumption was not constant over time (Da- browska-Prot et al. 1966, 1968). Other obser- vations have shown that spiders eat mosqui- toes in Japanese homes (Ori 1974) and in the prairies of Nebraska (Rapp 1978). Another ap- proach has been to test wild-caught spiders for the presence of mosquito antigen in their guts by the use of antibody-based tests. This meth- od showed that a large proportion of spiders in Kenyan homes were eating mosquito vec- tors of malaria (Service 1973). In Malaysia, spiders were eating Ae. albopictus (Skuse 1894), mosquito vectors of dengue outdoors (Sulaiman et al. 1990a), and Ae. aegypti vec- tors indoors (Sulaiman et al. 1990b). We saw only one study in which an attempt was made to determine whether spiders had a significant impact on a mosquito population in the field (Ramoska & Sweet 1981). That study found that discarded tires colonized by spiders con- tained fewer mosquito larvae than those tires without spiders. Since spiders generally eat a variety of prey, their usefulness as a biocon- trol agent of mosquitoes would depend on the relative abundance of mosquitoes and other prey species. Maximum effect from spiders could be expected where the microhabitat of the mosquito and the spider coincide so that a large proportion of prey are mosquitoes. While studying dengue virus transmission in rural Thailand, we noticed populations of Crossopriza lyoni (Blackwall 1867) (Araneae, Pholcidae) in village homes. The spiders were commonly seen in their webs, constmcted un- der homes and in undisturbed areas of the primitively constructed walls. Crossopriza lyoni generally inhabit the interiors of build- ings and other protected areas in southern Asia and Japan. Most of the literature on the species is restricted to taxonomic treatments (Yaginuma 1986; Kim 1988; Koh 1989) and studies on limited aspects of its physiology and behavior (Maya et al. 1982; Karuppa- swamy et al. 1984; Downes 1987). One Indian study (Nandi & Raut 1985) noted that C. lyoni eats Aedes species indoors. We suspected that the spiders could have an influence on adult populations of the dengue vector because the microhabitat of the spiders corresponded closely to the distribution of adult Ae. aegypti indoors. Predation on adult mosquitoes might be particularly significant in reducing transmission of dengue by Ae. ae- gypti, since the adult population includes older females which have survived long enough to acquire the virus and incubate it to infectious levels. In order to evaluate the possible role of C. lyoni in the ecology of dengue trans- mission, we made observations on bionomics of reproduction, development, and mosquito predation of the spider. In addition, we per- formed experiments to determine whether the spider might harbor dengue virus following feeding on an infected mosquito. METHODS Spiders. — Spiders were collected in and around homes of a village (official designation was Village 6) located 100 km east of Bang- kok in Hua Sam Rong District, Plaeng Yao County, Chachoengsao Province, Thailand. The spiders were captured incidentally during weekly sampling for Aedes aegypti (Edman et al. 1992; Scott et al. 1993). Not all spiders were retained and no attempt was made to quantify the abundance or variety of spider species. The spiders used in this study were perceived to be the most abundant kind during initial sampling. The pholcid specimens were identified as Crossopriza lyoni from the habitus, presence of depressed thoracic fovea, eye pattern, dis- tinct abdominal shape, morphology of the male left palpus, and morphology of the dis- sected and cleared female epigynum (Yagi- numa 1986; Kim 1986; Koh 1989). Voucher specimens of C. lyoni are deposited in the U.S. National Museum arachnid collection. The authors are confident of the identifica- tions, since the third author has taxonomic ex- perience with spiders and the specimens were carefully examined. It is possible that houses in the field also contained separate but mor- phologically similar species, because we did not perform a thorough survey of all spiders in the area. The work reported in this paper, however, was certainly performed on the stat- ed species, since specimens were examined from representative familial lines reared in the laboratory. The device for sampling was a commercial vacuum cleaner fitted with a screen-backed collection carton (12 cm diameter) affixed to a 0.5 m long section of PVC pipe. The pipe with the carton at the end was applied to crev- ices and spaces on the interior and exterior sides of the walls of the houses. Samples were 196 THE JOURNAL OF ARACHNOLOGY quickly chilled over wet ice and then refrig- erated at 4 °C overnight before sorting. Spiders were maintained in the laboratory at 30 °C (a representative temperature of the interior of village homes) and equal photo- phase and scotophase of 15 h. The spiders were kept in clear plastic cages (13 X 8 X 6.5 cm high) with tight lids. The lids were fitted with a small hole to introduce food and a 2-cm hole covered with screen for ventilation. The screen was covered with a square of gauze, which was wetted daily. Behavioral and developmental observa- tions.— Observations on behavior, feeding, and development were made during nine months on 13 different cages of spiders col- lected January-March 1991. In addition to general observations, quantitative measure- ments were made on growth of spiderlings and on rate of feeding by adult female spiders. Growth was observed by measurements of body length (chelicerae to posterior of abdo- men) twice per week, accomplished with the aid of a drawing tube attached to a dissecting microscope. The drawing tube was positioned over a digitizing tablet (Numonics Coip., Montgomeryville, Pennsylvania) and the length recorded by placing the pointing device of the tablet over the perceived image of a spiderling. Sigma Scan software (Jandel Sci- entific, Inc., Corte Madera, California) was used to calibrate the tablet and to record and analyze the data. Spiderlings were from a sin- gle egg sac, but were divided one day after hatching into a group of 25 in a cage by them- selves, and 24 in a cage with their mother. Data were analyzed with an independent t- test, comparing the difference between the mean lengths of spiderlings in the two cages each day that measurements were made. Throughout the 71 days of measurements, cages had constant access to an excess of Anopheles dims Peyton and Hamson 1979 mosquitoes for food. The number of mosquitoes eaten by female spiders was recorded for three individuals fed An. dims and for two individuals fed Ae. ae- gypti. Each day, the number of mosquitoes consumed by a spider was recorded and an excess of mosquitoes added to each cage. If all mosquitoes were consumed, a greater num- ber of mosquitoes was added the next day. Dengue virus experiment. — An experi- ment was performed to determine whether dengue virus in mosquitoes eaten by spiders could subsequently infect the spiders. Male Ae. aegypti mosquitoes were injected in the thorax with 0.017 p.1 of a tissue culture sus- pension of dengue 2 virus (10^ plaque-forming units (PFU)/ml) and then held at 32 °C for 10 days to allow time for the virus to amplify. In our laboratory, this procedure had been found to infect in excess of 90% of mosquitoes. Live infected mosquitoes (uninfected mosquitoes for controls) were fed to individually caged spiders which had been reared to maturity in the laboratory from eggs deposited by field- caught females. All spiders ate either three or four infected mosquitoes. After either 14 or 28 days, the spiders (one control and five vi- rus-fed spiders for each time interval) were dissected into three pieces which were sub- sequently kept cold over wet ice. The pieces were: 1) poison gland, prepared by cutting a wedge from between the first and second legs on each side to the area just behind the eyes, 2) prosoma, prepared from the remainder of the prosoma, and 3) abdomen. For the 14-day samples, the poison gland was triturated in 150 p.1, and the prosoma and abdomen each in 300 fjil of 20% fetal bovine serum in phos- phate buffered saline (FCS-PBS). For the 28-day sample, all parts were triturated in 500 |jl1 of FCS-PBS. Each triturate was injected into five Toxorhynchites splendens (Wiede- mann 1819) mosquitoes for amplification and detection of dengue virus (Rosen 1981). The number of poison gland triturates was limited to two virus-fed and one control spider for each time interval. The Tx. splendens mos- quitoes were held for 12 days at 30 °C before examining them for signs of infection using indirect immunofluorescent assay of head squashes (Sithiprasasna et al. 1994). In addi- tion, aliquots of the triturates were frozen at -70 °C until being tested using a double sand- wich enzyme-linked immunosorbent assay (ELISA) designed to detect dengue virus (Sithiprasasna et al. 1994). Controls were run with the ELISA to determine the sensitivity of the method used on spider tissue, triturating each tissue in 800 jxl of FCS-PBS. Five rep- licates of each control preparation were run, consisting of serial dilutions of dengue 2 seed (10^’ PFU/ml) diluted in either FCS-PBS, pre- viously-frozen spider triturates, or fresh spider triturates. STRICKMAN ET AL.—CROSSOPRIZA LYONI IN THAILAND 197 RESULTS Most spiders were collected from the inte- riors of homes between exposed support beams or behind furniture. Some spiders were also collected in the 1-3 m space under hous- es with elevated floors. The homes had wood- en floors, either wooden or bamboo walls, and metal or cement composite roofs. Construc- tion left many gaps in the walls and floors, forming holes that opened directly outdoors. Aedes aegypti mosquitoes were abundant in- doors (Edman et al. 1992; Scott et al. 1993) because of the open nature of the houses and because of the storage of large amounts of wa- ter for household use. Spiders copulated readily in the laboratory. In one case, a single male (70 days old, reared in the laboratory) copulated successfully with three females during a nine day period. The first time was observed immediately after the male was introduced into the cage of a female collected in the field 50 days before. The pair remained in copula for 40 minutes, with the female oriented ventral side up and the male facing her posterior with both palpi inserted into her genital orifice. The male was exposed to two other females for one day each, one collected 94 days and the other 116 days pre- viously. The third female ate the male, but ap- parently had copulated successfully. Fertile eggs were deposited 6, 4, and 5 days after copulation with each female, respectively. Oviposition was not observed directly, but resulted in an egg sac held in the mouthparts of the female, as is typical of the family. Cot- tony flecks in the web were observed 10 times in association with oviposition (occurring up to four days before and five days after) and six times not in association with oviposition. On one occasion, six eggs fell from an egg sac to the floor of the cage and subsequently did not hatch. The number of spiderlings hatching from 12 sacs deposited by nine spi- ders ranged from 5-54 with a mean of 34 (±SD == 14.8) spiderlings. Eggs failed to hatch only once. In most instances, hatching was noted when spiderlings were seen in the mother’s web, 11-13 days after oviposition. In one case, closer observation indicated that the spiderlings partially emerged from their eggs three days before they actually left the egg sac. The mother would hold onto the egg sac DAY NUMBER Figure 1. — Length of Crossopriza lyoni spider- lings emerging on Day 0 from a single egg sac, fed an excess of Anopheles dims mosquitoes, and mea- sured twice weekly. Twenty-four spiderlings re- mained in the cage with their mother (solid line) and 25 spiderlings were placed in a cage by them- selves (dashed line). Asterisks indicate days on which there was a statistically significant {P < 0.05) difference in length. until all spiderlings had left it, even when this process took more than one day. Spiderlings were inactive for the first 2-4 (mode = 3) days after leaving the egg sac, when molting occurred. Growth continued during the entire 7 1 days that spiderlings were measured (Fig. 1). Those spiderlings that had been left with their mother were consistently smaller than those spiderlings in a cage with- out an adult spider. This size difference was observed until the end of the measurement pe- riod, at which time there was no significant difference in size. The spiderlings separated from the mother matured more rapidly and de- posited their first egg sac when 74 days old, compared to 80 days for those spiders left with their mother. Mature females had a mean weight of 28.6 mg (n = 15, SD = 7.79, range: 18.2-44.6), 63% greater than the mean weight for males (17.6 mg, n = 9, SD == 3.73, range: 11.8-23.4). Although we did not hold labo- ratory-reared spiders long enough to get an 198 THE JOURNAL OF ARACHNOLOGY Table 1. — Number of mosquitoes (Anopheles dims or Aedes aegypti) eaten by mature, female, individ- ually-caged Crossopriza lyoni. Replicate spider #1 had male present 7 days; spider #4 had male present 1 1 days. Prey species “d” was An. dims, prey species “a” was Ae. aegypti. Replicate spider 1 2 3 4 5 Mean Prey species d d d a a No. of days 67 67 67 66 66 Spider weight (mg) 37 45 23 42 25 34.4 Mean eaten per day 1.6 0.81 1.4 0.89 0.73 1.1 SD eaten per day 1.3 0.87 1.4 1.1 0.85 1.2 Max. eaten per day 4 4 6 4 3 4.2 % days not eating 40 42 34 47 48 42.2 accurate estimate of longevity, we observed that wild-caught mature females lived as long as 120 days in the laboratory, implying lon- gevity of at least 194 days. Mature spiders captured mosquitoes which landed on their webs or which flew nearby. Hungry spiders actively pursued prey within their cages, generally capturing the mosquito within seconds of its introduction. The spider threw silk over the mosquito, the spider guid- ing the silk with its hind legs. The prey was then wrapped loosely in silk by manipulating the silk with abdomen and hind legs, but with- out rotating the prey. The first time a spider bit its prey was at the time of consumption, sometimes delayed as long as six days after capture. The quantity of mosquitoes con- sumed (Table 1) varied among individual spi- ders, but generally approached one mosquito per day, regardless of mosquito species. Feed- ing was discontinuous, with spiders fasting 34-48% of the days. Spiders with egg sacs continued to feed at approximately the same rate, setting aside the egg sac temporarily in order to consume the prey. Feces appeared as dark, taiTy spots on the floor of the cage. Spiderlings began feeding 2-4 days after their first molt, at which time they could over- power a mosquito which was 4.0 mm long (i.e., approximately 4X the length of the spi- der). Up to three spiderlings at once some- times fed on a single, wrapped mosquito. Spi- derlings sometimes fed on a mosquito wrapped by their mother or caught in their mother’s web. Three different cohorts of spi- derlings ate between 0.178-0.523 mosquitoes per spiderling per day during the first 11-17 days after beginning to eat. Cannibalism was common among the spiderlings, especially following introduction of mosquitoes when activity was at its greatest. Although probably an artifact of the confined conditions within a cage, cannibalism caused 67-84% mortality in four separate cohorts which were maintained until maturity. Toxorhynchites splendens were not infected by triturates from spiders which had fed on dengue-infected mosquitoes. Also, none of the spider triturates were positive for virus in the ELISA. The ELISA was sufficiently sensitive to detect a dilution of 1:160 (6.25 X 10"' PFU/ ml) of the virus seed in any of the fresh or frozen spider tissue triturates (Table 2). DISCUSSION Our laboratory observations on C. lyoni help fill in some of the gaps in bionomic knowledge of this species. Females deposited eggs shortly after copulation. The male was capable of mating successfully at least three times over a nine-day period, suggesting that a small number of males could keep a large group of females inseminated. Despite previ- ous reports (Downs 1987), we saw no evi- dence of the female eating any of her own eggs, possibly because most eggs were fertile. Prey-capturing techniques were described in detail by Nandi & Raut (1985), including ma- nipulation of silk and prey with the hind legs and biting only at the time of feeding. In ad- dition, they noted that the spiders actively re- moved carcasses of prey from the web. One of the interesting aspects of the spi- ders’ behavior was the interaction of the spi- derlings with their mother and with each other. The mother spider could evidently sense the presence of spiderlings in the egg sac, since the sac was retained until all spiderlings had STRICKMAN ET AU— CROSSOP RIZA LYONl IN THAILAND 199 Table 2. — ELISA sensitivity to dengue 2 virus (seed from tissue culture, 10^ PFU/ml) in Crossopriza lyoni tissue triturates. Diluent source Virus dilution n Mean optical density (O.D.) Poison gland Prosoma Abdomen Virus seed 1:2 5 0.550 0.347 0.447 1:8 5 0.301 0.193 0.298 1:16 5 0.224 0.164 0.207 1:32 5 0.179 0.128 0.154 Fresh spider 1:5 5 0.305 0.287 0.313 1:40 5 0.126 0.106 0.091 1:80 5 0.093 0.088 0.080 1:160 5 0.099 0.079 0.066 Frozen spider 1:5 5 0.291 0.350 0.312 1:40 5 0.115 0.108 0.096 1:80 5 0.098 0.087 0.079 1:160 5 0.080 0.082 0.068 Cutoff value 2 0.080 0.052 0.058 Infected Toxorhynchites 0.182 0.164 0.186 left it. The mother’s hunting activity some- times benefitted the spiderlings when they ate mosquitoes captured and wrapped by their mother or mosquitoes trapped in the mother’s web. Despite the apparent advantages near their mother, spiderlings kept by themselves grew and matured significantly faster than those kept with their mother, probably because they conserved energy which would have been spent following disturbance by the mother and because they were not competing with the mother for food. Among themselves, the spiderlings interacted in at least two ways. First, several spiderlings sometimes fed si- multaneously on the same mosquito. Second, the spiderlings ate each other, especially when excited by the introduction of prey. Such can- nibalism was a significant mortality factor in the confined conditions of a cage, though it was not observed in the field. We thought there was a possibility that spi- ders could harbor dengue virus, since spiders in village homes undoubtedly eat infected Ae. aegypti. Our laboratory experiment failed to demonstrate the presence of virus in spiders which had fed on dengue-infected mosquitoes. Triturates of the spiders were negative for vi- rus when injected into Toxorhync kites and when triturates were tested directly with an ELISA capable of detecting low titers of virus in spider tissues. Judging from observations of spiders feed- ing on mosquitoes in the laboratory, spiders could have a significant impact on the popu- lation of Ae. aegypti in a home. Our estimate of consumption was about one mosquito per mature female spider per day, but under other conditions this rate might be much higher. By feeding recently killed mosquitoes to C. lyoni occurring naturally in a house, Nandi & Raut (1985) observed that a single spider ate 12 — 20 mosquitoes per day for 2-3 consecutive days. Although we did not survey for other prey, small flies and spiders could have formed a part of the diet of C. lyoni in the field, diluting its effect on mosquitoes. It is significant, however, that juvenile and mature spiders were efficient at capturing mosquitoes and frequented the dark comers and walls of homes, corresponding to the locations favored by Ae. aegypti (Sheppard et al. 1969; Kusa- kabe & Ikeshoji 1990). Although we 'did not determine the number of spiders in village homes, all available microhabitats were usu- ally occupied. The potential significance of this predator raises the possibility that insec- ticidal application directed at Ae. aegypti adults indoors might actually exacerbate the dengue problem in mral Thailand. Indoor in- secticidal fogging might eliminate both mos- quitoes and spiders from inside a home, but the mosquitoes could quickly recolonize the house from existing larval sources. On the other hand, the spider population would re- 200 THE JOURNAL OF ARACHNOLOGY cover much more slowly because a greater proportion of the total population would have been exposed to insecticide and the spider’s reproductive rate is far lower than that of the mosquito (Christophers 1960). Crossopriza lyoni could prove valuable as an intentionally managed biocontrol agent for reduction of Ae. aegypti populations and den- gue transmission. There is precedence for the use of spiders to control a public health pest indoors, an example being the successful re- duction of fly populations and subsequent transmission of gastrointestinal pathogens (Nyffeler & Benz 1987). Introduction of C lyoni into homes without spiders could result in a constant population, self-regulated by cannibalism and availability of appropriate microhabitats. Because spiders eat a variety of prey, they would tend to maintain their pres- ence even when mosquitoes were scarce. As a result, spiders would be present to blunt sud- den mosquito population outbreaks (Riechert 1974). Such outbreaks can occur when rains fill many containers at once, hatching mos- quito eggs in all of them simultaneously. Where the spiders occur naturally, efforts could be made to avoid killing spiders during housecleaning or insecticidal application. The presence of dengue where the spiders now oc- cur shows that spiders alone do not stop trans- mission; however, management of spider pop- ulations might provide the additional control of adult mosquitoes needed to block dengue transmission following reduction of larval populations by other, non-insecticidal means (e.g., Kittayapong & Strickman 1993). LITERATURE CITED Christophers, S.R. 1960. Aedes aegypti (L.), the yellow fever mosquito. Its life history, bionom- ics, and structure. Cambridge Univ. Press, Cam- bridge. 739 pp. Dabrowska-Prot, E., J. Luczak, & K. Tarwid, 1966. Experimental studies on the reduction of mos- quitoes by spiders. III. Indices of prey reduction and some controlling factors. Bull. Acad. Polo- naise Sci., Ser. Sci. Biol., 14:777-782. Dabrowska-Prot, E., J. Luczak & K. Tarwid, 1968. The predation of spiders on forest mosquitoes in field experiments. J. Med. EntomoL, 5:252-256. Downes, M.E 1987. Crossopriza (lyonil) (Arane- ae, Pholcidae) eats her own eggs. J. Arachnol., 15:276. Edman, J.D., D. Strickman, P. Kittayapong & TA. Scott. 1992. Aedes aegypti (Diptera: Cu- licidae) in Thailand rarely feed on sugar. J. Med. EntomoL, 29:1035-1038. Gubler, D.J. 1988. Chapter 23. Dengue. Pp. 223- 260. In The arboviruses: Epidemiology and ecol- ogy, Volume II, (TP. Monath, ed.). CRC Press, Boca Raton, Florida. Karuppaswamy, S.A., M. Maya & K. Palanisamy. 1984. Effects of chemicals on the web of the spiders Hippasa greenalliae (Blackwall) (Lycos- idae) and of Crossopriza lyoni (Blackwall) (Pholcidae). Comp. Physiol. Ecol., 9:275-276. Kim, J.P. 1988. One species of genus Crossopriza (Araneae: Pholcidae) from southern Asia. Kore- an J. Arachnol., 4:35-38. Kittayapong, P. & D. Strickman. 1993. Three sim- ple devices for preventing development of Aedes aegypti larvae in water jars. American J. Trop. Med. Hyg., 49:158-165. Koh, J.K.H. 1989. Guide to common Singapore spiders. Singapore Science Centre, Singapore, 160 pp. Kusakabe, Y. & T. Ikeshoji. 1990. Comparative at- tractancy of physical and chemical stimuli to ae- dine mosquitoes. Japanese J. Sank. Zool., 41: 219-225. Maya, M., M. Palaniswamy & S.A. Karuppaswamy. 1982. A study on the contact sex pheromone in Crossopriza lyoni (Blackwall) (Pholcidae). Comp. Physiol. Ecol., 7:22-24. Nandi, N.C. & S.K. Raut. 1985 (1986). Predatory behaviour of the pholcid spider Crossopriza lyoni (Blackwall) on mosquitoes {Aedes sp.). Bull. Zool. Surv. India, 7:179-183. Nyffeler, M. & G. Benz. 1987. Spiders in natural pest control: A review. Zeitschrift fiir angewand- te, 103:321-339. Ori, M. 1974. Studies on spiders as natural ene- mies of insect pests. 1. Observations on the spi- ders in houses in Nagasaki Prefecture. Japanese J. Sank. Zool., 25:153-160. Ramoska, W.A. & R.A. Sweet. 1981. Predation on mosquitoes (Diptera: Culicidae) breeding in tires by the spider Agelenopsis naevia (Araneae: Age- lenidae). J. Med. EntomoL, 18:355-356. Rapp, W.E 1978. Preliminary studies of Tetra- gnatha (Araneida) as predators of mosquitoes. Mosquito News, 38:506-507. Riechert, S.E. 1974. Thoughts on the ecological significance of spiders. Bioscience, 24:352-356. Rosen, L. 1981. The use of Toxorhynchites mos- quitoes to detect and propagate dengue and other arboviruses. American J. Trop. Med. Hyg., 30: 177-183. Scott, TW., E. Chow, D. Strickman, P. Kittayapong, R.A. Wirtz, L.H. Lorenz & J.D. Edman. 1993. Blood-feeding patterns of Aedes aegypti (Dip- tera: Culicidae) collected in a rural Thai village. J. Med. EntomoL, 30:922-927. Service, M.W. 1973. Identification of predators of STRICKMAN ET Al.,— CROSSOP RIZA LYONI IN THAILAND 201 Anopheles gambiae resting in huts, by the pre~ cipitin test. Trans, R. Soc. Trop. Med. Hyg., 67: 33-34. Sheppard, RM., W.W. Macdonald, R.J. Tonn & B. Grab. 1969. The dynamics of an adult popula- tion of Aedes aegypti in relation to dengue hae- morrhagic fever in Bangkok. J. Aeim. EcoL, 38: 661-702. Sithiprasasna, R., D. Strickman, B.L. Innis & K.J. Linthicum. 1994. ELISA for detecting dengue and Japanese encephalitis viral antigen in mos- quitoes. Ann. Parasitol. Trop. Med., 88:397-404. Sulaiman, S., B. Omar, S. Omar, I. Ghauth & J. Jeffery. 1990a. Detection of the predators of Ae- des albopictus (L.) (Diptera: Culicidae) by the precipitin test. Mosquito-Bome Diseases Bull., 7: 1-4. Sulaiman, S., B. Omar, S. Omar, I. Ghauth & J. Jeffery. 1990b. Detection of Aedes aegypti (L.) (Diptera: Culicidae) predators in urban slums in Malaysia using the precipitin test. Mosquito- Borne Diseases Bull,, 7:123-126. Yaginuma, T. 1986. Spiders of Japan in color. New edition. Hoikusha Publishing Co., Ltd., Osaka, Japan. 305 pp. Manuscript received 21 March 1996, accepted 4 March 1997. 1997. The Journal of Arachnology 25:202-205 LEG AUTOTOMY AND AVOIDANCE BEHAVIOR IN RESPONSE TO A PREDATOR IN THE WOLF SPIDER, SCHIZOCOSA AVIDA (ARANEAE, LYCOSIDAE) Fred Punzo: Department of Biology, University of Tampa, Tampa, Florida 33606 USA ABSTRACT. The wolf spider, Schizocosa avida (Walckenaer 1838) can utilize leg autotomy to suc- cessfully avoid capture by the scorpion, Centruroides vittatus (Say 1821). Leg autotomy was used suc- cessfully in 7 out of 43 encounters with a scorpion (16%). Spiders were captured and eaten by scorpions in 79% of the encounters. Two spiders (5%) escaped capture by means other than leg autotomy. Scorpions most often grasped spiders at their abdomen (49%), followed by the cephalothorax (35%) and leg (16%). Naive spiders (no previous experience with a scorpion), with intact legs or an autotomized leg, spent significantly more time (26-32 min out of a 60-min trial) on a filter paper disc that had come into previous contact with a scorpion as compared to spiders that had lost a leg in a successful escape from an encounter with the predator (10 min). This is an example of associative avoidance learning and is the first demon- stration of this type of learning in response to previous experience with a predator in spiders. The autotomy of various bodily structures in response to attack by predators has been reported for many species (Robinson et al. 1970; Edmunds 1974). Tail autotomy has been shown to enhance survival in salamanders (Brodie 1983) and lizards (Punzo 1982; Ar- nold 1988). Decapod crustaceans (Spiviak & Politis 1989) and spiders (Foelix 1982; For- manowicz 1990) frequently autotomize their legs when grasped by predators. In the present study, I analyzed the effec- tiveness of leg autotomy as an antipredator strategy in the wolf spider, Schizocosa avida (Walckenaer 1838), against a naturally occur- ring scorpion predator, Centruroides vittatus (family Buthidae). I also investigated the ef- fects of previous encounters with the predator on the avoidance behavior of S. avida. METHODS I collected adult females of S. avida in Brewster County, Texas, during May-July 1994. Spiders carrying egg sacs were located at night using a headlamp. Scorpions (C. vit- tatus) from the same ai'ea were located using a portable UV light (BioQuip Model 2813C, Gardena, California). Animals were placed in- dividually in plastic holding containers and transported to the laboratory for subsequent studies. Voucher specimens of S. avida and C vittatus have been deposited in the University of Tampa Invertebrate Collection. All experiments were conducted on adult female spiders (body length: 11-14 mm) reared from egg cases collected in the field. Spiderlings emerging from egg cases were reared in an environmental chamber (Percival Model 1-37, Boone, Iowa) maintained at 20 °C, 68-72% relative humidity (RH), and a 12L:12D photoperiod regime. Spiderlings were housed individually in plastic containers and fed on a mixed diet of flies {Drosophila melanogaster and D. virilis) and cockroach {Periplaneta americana) nymphs. Water was provided ad libitum. As the spiders grew in size, larger prey were used (adult crickets, cocki*oaches and beetles). As a result of these rearing procedures, all spiders were naive in the sense that they had no prior experience with the scorpion, Centruroides vittatus, which is sympatric with Schizocosa avida in Brewster County, Texas. Adult female scor- pions (0.419 g ±0.21) were maintained under the same conditions and fed on a diet of crick- ets, grasshoppers and a variety of spiders col- lected locally in Hillsborough County, Flori- da. Scorpions were deprived of food for 72 h prior to any encounter with a spider. Encounter experiments. — Each encounter between a scorpion and a spider was con- ducted in a clear acrylic plastic (Plexiglass®) chamber (15 X 10 X 10 cm) at room temper- ature. The chamber was situated on a wooden 202 PUNZO— WOLF SPIDER AVOIDANCE BEHAVIOR 203 table behind a one-way mirror to minimize disturbance to the animals during encounter sequences. The floor of the chamber contained 3 cm of sand as a substrate. All spiders used in these experiments were fed 24 h prior to an encounter with a predator, and possessed all of their legs. A scorpion was placed in the chamber for 24 h prior to an encounter with a spider. I staged a total of 50 encounters (tri- als) between different spiders and scorpions. Each spider and scorpion was tested only once. At the start of each trial, a spider was placed at the center of the encounter chamber which contained one randomly chosen scor- pion. I carefully observed both animals and verbally recorded their activities using a Sony HP- 110 tape recorder. In seven of the trials, the scorpions made no attempt to capture the spider. Only those data obtained from trials in which a scorpion attempted to capture a spider {n = 43) were used for statistical analysis. A trial ended when the spider was either suc- cessfully captured and ingested by a scorpion, escaped an initial strike without utilizing leg autotomy, or escaped via leg autotomy. Data were analyzed using the G-test of indepen- dence as described by Sokal & Rohlf (1981). All encounters were recorded on a Panasonic L3 video recorder for subsequent study as pre- viously described by Punzo (1995). Effect of previous encounter. — In a sec- ond set of experiments using different spiders and scorpions, I tested the effects of a previ- ous encounter with C. vittatus on the subse- quent behavior of three groups of S. avida. One group of spiders (Gl, n = 15) consisted of individuals who had all of their legs intact and had never encounterd a scorpion through- out their lives. Another group (G2) consisted of 15 different spiders who had also never en- countered a scorpion; in this group, however, one of their legs (chosen at random) was au- totomized after being grasped by a pair of for- ceps. The third group (G3) consisted of 15 spiders who had one previous experience with a scorpion and had used leg autotomy to suc- cessfully escape capture by the predator. In these experiments, a spider was placed ran- domly into either end of a glass chamber (15 X 15 X 8 cm) containing two, square-shaped pieces of filter paper (Whatman No. 1) situ- ated side-by-side and covering the entire floor of the chamber. One of the pieces of filter pa- per was taken from the floor of a plastic con- tainer housing a scorpion (treated), allowing C vittatus to come into contact repeatedly with the filter paper during the course of its normal activities (for a period of two days). The other fresh piece of filter paper (untreat- ed) had not been in contact with a scorpion. For each trial, the positions of the two pieces of filter paper (to the right or left) on the floor of the chamber was determined using a table of random numbers. The length of each trial was 60 min, and the amount of time (in min) spent by each spider on both pieces of filter paper was recorded with a stopwatch. A Dun- can multiple range test (Sokal & Rohlf 1981) was used to analyze the data. RESULTS In encounters with scorpions, S. avida fe- males were successfully captured and eaten in 79% of the trials (34 out of 43 trials). Nine spiders escaped capture; seven of these (78%) utilized leg autotomy. During prey capture, scorpions grasped the spider using one of their pedipalps. Spiders were either grasped by their abdomen (49%), cephalothorax (35%) or leg (16%). Only two of the nine spiders (22%) escaped capture without utilizing leg autoto- my. These two spiders were grasped at the distal end of the abdomen and lateral region of the cephalothorax and escaped by pulling free from the pedipalps. Sixteen percent of the spiders were grasped by a leg and escaped af- ter autotomizing the limb. Significantly more spiders escaped capture by utilizing leg autot- omy than those who escaped by struggling free (G = 34.9, df = I, P < 0.001). Scorpions were observed feeding on the autotomized leg while the spider ran to the other end of the chamber. The effects of previous encounters with a predator on the subsequent behavior of S. av- ida are shown in Figure 1. There was no sig- nificant difference in the mean amount of time spent on a treated (contact with scorpion) ver- sus untreated (no contact with scorpion) piec- es of filter paper between G1 (legs intact, no previous encounter with predator) and G2 (leg autotomized, no previous experience) spiders (P > 0.50). However, spiders that had previ- ously escaped by autotomizing a leg (G3) spent significantly less time on treated filter paper (10 min out of a 60 min trial) as com- pared to untreated filter paper (50 min) (P < 0.001). 204 THE JOURNAL OF ARACHNOLOGY 50 45 CO 40 Lu 35 h30 Z)25 Z20 - 15 ^10 5 0 TREATED UNTREATED Figure 1 . — The effects of previous encounter experience with the scorpion Centruroides vittatus on the behavior of three groups (Gl, G2, G3) of Schizocosa avida. Values represent the mean amount of time (in minutes) that Schizocosa avida females remained on square pieces of filter paper that had been exposed to the presence of Centruroides vittatus (treated) as compared to discs that had not been contacted by the scorpion (untreated). Gl = spiders with legs intact, no previous experience with a scorpion; G2 = leg autotomized, no previous experience with a scorpion; G3 = leg autotomized in an encounter with a scorpion. Vertical lines represent SD. See text for details. DISCUSSION This study demonstrates that S. avida can utilize leg autotomy to escape capture by a natural predator if grasped by the leg. A pre- vious study by Formanowicz (1990) showed that a filistatid spider, Kukulcania hibernalis (Chamberlin 1926) from Wise County, Texas, was also able to utilize leg autotomy to escape predation by C.entruroides vittatus. However, this strategy was not an effective defense against a centipede predator (Scolopendra polymorpha). In the present study, leg autot- omy resulted in a successful escape in 16% of the encounters for the wolf spider S. avida, whereas K. hibernalis successfully utilized this strategy in 36% of its encounters with a scorpion. The site of autotomy was always at the junction (intersegmental membrane) between the coxa and trochanter of the leg grasped by the scorpion. This is in agreement with the previous literature on leg autotomy in spiders (Robinson et al. 1970; Foelix 1982) and some insects (Pearson 1985). In all cases, once the spider was grasped by a scorpion, it exhibited a rapid upward movement of the coxa. The rest of the distal portion of the leg remained in a relatively fixed in position. This study also shows that once S. avida has had an encounter experience with C. vit- tatus and is successful in escaping capture, it will avoid a substrate that has been previously occupied by this scorpion. This suggests that the spider can remember some cue (probabaly olfactory in nature) associated with the scor- pion and use that information to avoid the predator at a later time. This is an example of rapid associative avoidance learning (Punzo 1985, 1996) and represents the first demon- stration that a spider can utilize this type of behavioral plasticity to avoid predators. Al- though this spider and scorpion have presum- ably coexisted sympatrically for a long period of time, there is no indication that S. avida possesses an innate capacity to recognize the presence of this predator. Spiders that had no previous encounter experience with the scor- pion did not demonstrate avoidance of a sub- strate previously occupied by C vittatus. ACKNOWLEDGMENTS I thank T. Punzo for assistance in field col- lecting, A. Jenzarli for statistical consultation. PUNZO—WOLF SPIDER AVOIDANCE BEHAVIOR 205 and C. Bradford, J, Berry, P. Sierwald, and anonymous reviewers for reading an earlier draft of this manuscript. I also thank the Uni- versity of Tampa for a Faculty Development Grant which made much of this work possi- ble. LITERATURE CITED Arnold, E.N. 1988. Caudal autotomy as a defense. Pp. 237-273, In Biology of the reptilia. Vol. 16 (C. Cans & R.B. Huey, eds.). Alan Liss, New York. Brodie, E.D. 1983. Antipredator adaptations of sal- amanders: evolution and convergence among ter- restrial species. Pp. 109-133, In Plant, animal and microbial adaptations to terrestrial environ- ment. (N.S. Margaris & R.J. Reiter, eds.). Ple- num Press, New York. Edmunds, M. 1974. Defense in animals. Longman, London. Foelix, R.F. 1982. Biology of spiders. Harvard Univ. Press, Cambridge, Massachusetts. Formanowicz, D.R. 1990. Antipredator efficcy of spider leg autotomy. Anim. Behav., 40:400-401. Pearson, D.L. 1985. The function of multiple an- tipredator mechanisms in adult tiger beetles (Co- leoptera: Cicindelidae). Econ. Entomol., 10:65- 72. Punzo, F 1982. Tail autotomy and running speed in the lizards, Cophosaurus texanus and Uma no- tata. J. HerpetoL, 16:329-331. Punzo, F. 1985. Recent advances in behavioral plasticity in insects and decapod crustaceans. Florida Entomol., 68:89-104. Punzo, F. 1995. The biology of the spider wasp, Pepsis thisbe (Hymenoptera: Pompilidae) from Trans Pecos Texas. I. Adult morphometries, lar- val development and the ontogeny of larval feed- ing patterns. Psyche, 101:229-241. Punzo, F. 1996. Localization of brain function and neurochemical events associated with learning in insects. Recent Trends in Comp. Biochem. Phys- iol. (in press). Robinson, M.H., L.G. Abele & B. Robinson. 1970. Attack autotomy: a defense against predation. Science, 169:300-301. Sokal, R.R. & F.J. Rohlf. 1981. Biometry. 2d ed. W.H. Freeman, San Francisco, California. Spiviak, E.D. & M.A. Politis. 1989. High inci- dence of limb autotomy in a crab population from a coastal lagoon in the province of Buenos Aires, Argentina. Canadian J. ZooL, 67:1976- 1985. Manuscript received 19 June 1996, accepted 10 November 1996. 1997. The Journal of Arachnology 25:206-212 TUBEROCHERNES (PSEUDOSCORPIONIDA, CHERNETIDAE), A NEW GENUS WITH SPECIES IN CAVES IN CALIFORNIA AND ARIZONA William B. Muchmore: Department of Biology, University of Rochester, Rochester, New York 14627 USA ABSTRACT. A new genus, Tuberochernes, is defined, the type species being T. aalbui new species, from a cave in Mono County, California. Another species referable to the genus is described also, T. ubicki new species, from a cave in Santa Cruz County, Arizona. Unique modifications of the palpal chelae and first legs of the males are discussed. Among some pseudoscoipions collected in caves in California and Arizona by Rolf Aal- bu, Derham Giuliani, and Darrell Ubick were some specimens with striking modifications of the palpal chelae and first legs of males. The combination of these features, unique among pseudoscorpions, warrants the description of two new species and the establishment of a new genus. METHODS Animals were collected in pitfall traps con- taining ethylene glycol or directly into alco- hol, and were stored in alcohol. Specimens were prepared for detailed study by dissection, clearing in clove oil, and mounting in Canada balsam on microscope slides, generally fol- lowing the procedure described by Hoff (1949). They were studied and measured un- der a compound microscope, and drawings were made by direct projection of the image onto paper. The specimens are deposited, as indicated, in the California Academy of Sci- ences, San Francisco, California (CAS) and the Florida State Collection of Arthropods, Gainesville, Florida (FSCA). A few abbreviations are used in the text: L = length; L/B = ratio, length/breadth; L/D = ratio, length/depth; T = tactile seta. Family Chernetidae Menge Chemetidae Menge 1855:22; Muchmore 1982:101; Harvey 1991:534 (complete synonymy to 1988); Harvey 1992:1427. Tuberochernes new genus Type species. — Tuberochernes aalbui Muchmore new species. Diagnosis. — Tuberochernes is unique among known chemetid pseudoscorpions in the possession of the following suite of char- acters: 1) cheliceral flagellum of four setae; 2) spermathecae of female consist of two long tubes with terminal sacs; 3) male (but not fe- male) with a conspicuous conical protuber- ance on the medial side of the hand of the palpal chela; 4) tarsus of leg I of male (but not female) distinctively shortened and curved; 5) tarsus of leg IV (both sexes) with a short, distally located, tactile seta, variably acuminate or finely denticulate. It appears most closely related to Mirochernes Beier 1930, from which it can be distinguished readily by the much larger and more complex process on the chelal hand of the latter. Description. — A genus of the family Cher- netidae. Palps well sclerotized, reddish-brown, carapace light brown, other parts lighter. Sur- faces of carapace and palps heavily granulate, with slender clavodentate setae. Carapace with two distinct, transverse furrows; without eyes; with 70-100 setae, four at anterior and 12-20 at posterior margin. Most tergites and stemites divided; middle tergites with 25-30 and ster- nites with about 20 setae. Cheliceral hand usu- ally with six setae (occasionally five), bs, sbs and X.S denticulate, es short and either acu- minate or finely denticulate; flagellum of four setae, the two distal ones long, serrate, the two proximal ones short, simple; galea slender, with 4-6 small rami. Palp rather slender, sex- ually dimoiphic; trochanter, femur and patella of both sexes, and chela of female, typically chernetid in proportions; chela of male heavi- 206 MUCHMORE— NEW PSEUDOSCORPIONS IN CAVES 207 er than that of female, with a large, broad- based, conical protuberance on medial side of hand, and with movable finger distinctly bowed. Chelal fingers with 45-50 cusped mar- ginal teeth, and with many external and a few internal accessory teeth. Venom apparatus de- veloped only in movable finger. Trichobothria as shown in Fig. 4; on fixed finger, ist well distad of est, and ib distad of esb; on movable finger, st distinctly closer to t than to sb. Leg I sexually dimorphic, especially in T. ubicki new species; segments more robust in male than in female; tarsus of male curved, while that of female straight. Leg IV quite slender. Tarsus IV with a short, distally located, tactile seta; this seta is sometimes acuminate, some- times denticulate at tip. Male genitalia typi- cally chemetid in form; spermathecae of fe- male are thin tubes with irregular, ovoid end sacs. Two species are presently assigned to Tu- berochernes, namely T. aalbui new species from Mono County California, and T. ubicki new species from Santa Cruz County, Arizo- na, as described below. Tuberochernes aalbui new species Figs. 1-8 Type data.— Holotype male (WM7269. 02005), female paratype (allotype) (WM7269. 02008), and about 180 other paratypes (in- cluding all stages) taken in ethylene glycol pitfall traps in Poleta Cave, Westgard Pass, Inyo-White Mountains, Inyo County, Califor- nia, 21 May-5 November 1988, R. Aalbu; ho- lotype, allotype and 28 paratypes (all stages) mounted on slides. Three paratypes (29,1 tri- tonymph) taken in antifreeze pitfall trap in Po- leta Cave (as “Westergard Pass Cave”), 27 May-26 November 1982, D. Giuliani; mounted on slides. All mounted specimens in FSCA, some alcoholic specimens in CAS. Diagnosis. — Much like T. ubicki new spe- cies, but specimens larger, palp and leg IV less stout. Tarsus of leg I in T. aalbui new species male more similar to that of conspecific fe- male than is the case in T. ubicki. Description. — Male and female similar in most respects, but female usually a little larg- er, and palpal chelae and first legs sexually dimorphic. Palps reddish-brown, carapace light brown, chelicerae and legs tan. Carapace longer than broad; surface granulate, with two distinct transverse furrows; no eyes; about 80 clavodentate setae, 4 at anterior and 12-16 at posterior margin. Abdominal tergites 1-10 and sternites 4-10 divided; surfaces of tergites lightly granulate; pleural membranes irregu- larly longitudinally striate; most dorsal setae slender and clavodentate, ventral setae very slender and denticulate or clavodentate. Tergal chaetotaxy of holotype 18:23:21:25:28:26:25: 24:24:22:T12T:2, others similar; tactile setae (T) apparently very fragile, as they are usually missing from their areoles. Sternal chaetotaxy of holotype (male) ~50:[33]:(2)29(1):(2)6(2): 16:21:21:21:21:18:T1T2T1T:2, other males similar; anterior chaetotaxy of allotype (fe- male) 32:(2)14(2):(3)6(4): 14:21:-, other fe- males similar. Internal genitalia of male typically chemetid in form, fairly large and well sclerotized. Spermathecae of female are long tubes with irregular, ovoid end sacs (Fig. 1); the tubes must be thin- walled and fragile, as none could be followed to a medial atrium. Chelicera 0.3 as long as carapace; hand usu- ally with 6 setae (occasionally 5), bs, sbs, and xs terminally denticulate, others acuminate; flagellum of 4 setae, the distal 2 long and an- teriorly serrate, the proximal 2 short and ter- minally denticulate; galea slender, with 5-6 small rami, equally developed in male and fe- male. Palp (Figs. 2, 3) rather long, stouter in male than in female: (numbers for male fol- lowed in parentheses by those for female): fe- mur 0,85-0.92 (0. 9-1.0) and chela about 1.25 ( 1.35-1. 45) X as long as carapace; L/B of tro- chanter 1.65-1.95 (1.85-2.15), femur 3.0- 3.25 (3.45-3.65), patella 2.65-2.8 (2.8-3. 1), and chela (without pedicel) 2.5-2. 8 (2.8-3. 1); L/D of hand (without pedicel) 1.4-1.55 (1.5- 1.7); movable finger L/hand L 0.95-1.1. Chela of male quite robust, with a conspicuous, con- ical protuberance on medial side of hand, and with movable finger distinctly bowed (Fig. 4); chela of female more slender and without these features. Surfaces lightly granulate; most setae narrow clavodentate. Trichobothria as shown in Figs. 4, 5. Each finger with 45- 50 contiguous, cusped marginal teeth; fixed finger with 9-11 external and 3-6 internal, and movable finger with 5-10 external and 1- 2 internal accessory teeth. Venom apparatus present only in movable finger, nodus ramosus between trichobothria t and st. Legs slender: leg IV (Fig. 6) with L/D of femur + patella 4.4-4.85, tibia 7. 0-7. 8, and tarsus 6.35-7.45. Leg I of male (Fig. 7) with tarsus slightly 208 THE JOURNAL OF ARACHNOLOGY Figures 1-8. — Tuberochernes aalbui new species. 1, Spermathecae of allotype female; 2, Right palp of holotype male, dorsal view; 3, Right palp of allotype female, dorsal view; 4, Left chela of paratype male, lateral view, showing trichobothriotaxy (all setae omitted; darkened areoles are underneath); 5, Left chela of allotype female; 6, Leg IV of holotype male (vestitural setae omitted); 7, Leg I of holotype male. 8, Leg I of allotype female. Scale bar = 0.15 mm for Fig. 1, and 0.5 mm for all others. bowed, concave dorsally; that of female (Fig. 8) normal, straight. Leg IV with a short tactile seta on tarsus 0.75-0.8 length of segment from proximal end; these setae are variably acuminate or terminally denticulate. Nymphs: Generally similar to adults, but progressively smaller, lighter in color, and with appendages slightly more robust. Hand of chelicera with fewer setae: tritonymph with 5 or 6 (xs sometimes absent), deutonymph MUCHMORE— NEW PSEUDOSCORPIONS IN CAVES 209 with 5 (xs absent), and protonymph with 4 (xs and bs absent); flagellum with 4 setae in all stages. Palp: fixed and movable fingers bear the typical numbers of trichobothria for each stage. Leg IV: tarsus bears a short, acuminate tactile seta 0.65-0.7 length of segment from proximal end. Measurements. — Male: Figures given first for holotype, followed in parentheses by rang- es for seven paratypes. Body L 4.22 (3.85- 4.68). Carapace L 1.30 (1.21-1.35). Chelicera L 0.37 (0.37-0.42). Palp: trochanter 0.63 (0.62-0.695)70.34 (0.325-0.38); femur 1.13 (1.06-1.23)70.355 (0.34-0.41); patella 1.01 (1.00-1.11)70.37 (0.37-0.415); chela (without pedicel) 1.61 (1.56-1.70)70.59 (0.555-0.665); hand (without pedicel) 0.815 (0.82-0.92)7 0.555 (0.53-0.62); pedicel L 0.15 (0.13-0.16); movable finger L 0.89 (0.83-0.925). Leg I: femur + patella L 0.83 (0.815-0.90); femur 0.455 (0.43-0.48)70.235 (0.235-0.265); patel- la 0.58 (0.56-0.62)70.19 (0.19-0.23); tibia 0.67 (0.64-0.75)70.13 (0.14-0.16); tarsus 0.615 (0.57-0.70)70.105 (0.09-0.11). Leg IV: femur+patella 1.02 (0.97-1.10)70.23 (0.215- 0.245); tibia 0.92 (0.865-1.02)70.125 (0.125- 0.14); tarsus 0.665 (0.615-0.74)70.105 (0.095-0.11). Female: Figures given first for allotype, fol- lowed in parentheses by ranges for 12 para- types. Body L 4.40 (3.65-4.85). Carapace L 1.29 (1.16-1.34). Chelicera L 0.39 (0.355- 0.41). Palp: trochanter 0.665 (0.615-0.69)7 0.32 (0.29-0.34); femur 1.18 (1.07-1.24)7 0.34 (0.30-0.36); patella 1.06 (0.955-1.11)7 0.38 (0.34-0.415); chela (without pedicel) 1.85 (1.61-1.88)70.62 (0.525-0.635); hand (without pedicel) 0.95 (0.83-1.01)70.585 (0.51-0.62); pedicel L 0.13 (0.11-0.13); mov- able finger L 0.93 (0.83-0.955). Leg I: fe- mur+patella L 0.835 (0.74-0.87); femur 0.42 (0.36-0.45)70.215 (0.185-0.22); patella 0.56 (0.52-0.605)70.18 (0.16-0.19); tibia 0.65 (0.58-0.70)70.125 (0.11-0.125); tarsus 0.63 (0.58-0.69)70.095 (0.09-0.095). Leg IV: femur+patella 1.07 (0.985-1.13)70.22 (0.205- 0.245); tibia 0.955 (0.865-1.04)70.13 (0.11- 0.13); tarsus 0.70 (0.635-0.725)70.095 (0.09- 0.105). Tritonymph: Ranges for five paratypes. Body L 2.85-3.30. Carapace L 0.89-1.05. Chelicera L 0.28-0.325, Palp: trochanter 0.445-0.5270.22-0.27; femur 0.725-0.897 0.235-0.29; patella 0.635-0.78570.265-0.325; chela (without pedicel) 1.23-1.4370.435-0.50; hand (without pedicel) 0.63-0.7370.43-0.495; pedicel L 0.08-0.09; movable finger L 0.665- 0.74. Leg IV: femur+patella 0.69-0.8270.16- 0.19; tibia 0.57-0.66570.105-0.12; tarsus 0.45-0.52570.08-0.09. Deutonymph: Ranges for three paratypes. Body L 1.95-2.25. Carapace L 0.615-0.67. Chelicera L 0.20-0.25. Palp: trochanter 0.30- 0.3570.15-0.18; femur 0.47-0.5570.15-0.185; patella 0.42-0,4870.17-0.215; chela (without pedicel) 0.83-0.9670.26-0.325; hand (without pedicel) 0.415-0.49570.26-0.31; pedicel L 0.05-0.07; movable finger L 0.46-0.495, Leg IV: femur+patella 0.445-0.52570.12-0.135. Protonymph: Ranges for three paratypes. Body L 1.45-1.75. Carapace L 0.48-0.56. Chelicera L 0.185-0.19. Palp: trochanter 0.235-0.24570.12-0.125; femur 0.34-0.3557 0.115-0.125; patella 0.3170.13-0.14; chela (without pedicel) 0.61 5-0.6570. 18--0. 20; hand (without pedicel) 0.32-0.3370.18-0.19; pedi- cel L 0.04; movable finger L 0.325-0.35. Etymology.— The species is named in hon- or of Rolf Aalbu, who collected most of the type specimens. Remarks. — The first legs of the male are somewhat modified compared to those of the female, though not so much so as in T. ubicki (see below). Tuberochernes ubicki new species Figs. 9-14 Type data. — Holotype male (WM7729. 01001) and female paratype (allotype) (WM7729.01002) from under stones in Fly Cave, Gardner Canyon, Santa Rita Mountains, Santa Cruz County, Arizona, 24 June 1988, D. Ubick; mounted on slides, in CAS. Diagnosis. — Much like T. aalbui, but a lit- tle smaller, with palp and leg IV a little stouter, and leg I of male apparently raptorial, the segments being distinctly modified. Description.— Male and female generally similar, but female a little larger, and palpal chelae and first legs sexually dimorphic. Palps reddish-brown, carapace light brown, chelic- erae and legs tan, other parts lighter. Carapace longer than broad; surface covered with low granules and with two distinct, transverse fur- rows; no eyes; about 90-100 clavodentate se- tae, 4 at anterior and 1 8-20 at posterior margin. Abdominal tergites 2-10 and stemites 4-10 di- vided; surface of tergites lightly granulate; 210 THE JOURNAL OF ARACHNOLOGY Figures 9-14. — Tuberochernes ubicki new species. 9, Right palp of holotype male, dorsal view; 10, Right palp of allotype female, dorsal view; 11, Left chela of holotype male, lateral view, showing tricho- bothriotaxy (all setae omitted; darkened areoles are underneath); 12, Leg I of holotype male (vestitural setae omitted); 13, Leg I of allotype female; 14, Leg IV of holotype male. Scale bar = 0.5 mm. pleural membranes inegularly longitudinally striate; most dorsal setae slender, clavodentate, ventral setae slender, acuminate to clavoden- tate. Tergal chaetotaxy of holotype 20:28:26: 28:27:29:30:26:23:21:T12T:2. Sternal chaeto- taxy of holotype (male) 60: [4-4] :( 1)25(1): (1)6(1): 15:20:20:20:20:18:T2T2T1T:2; anterior chaetotaxy of allotype (female) 34: (1)9(1): (1)5(1): 16:22:-. Internal genitalia of male typ- ically chemetid in form; spermathecae of fe- male not clear, but apparently like those of T. aalbui new species. Chelicera 0.25 as long as carapace; hand with 6 setae, is and Is long, acuminate, others rather short and terminally denticulate; flagellum of 4 setae, distal 2 long and serrate, proximal 2 short and denticulate near tip; galea of male small, with 3-4 spi- nules, that of female longer, slender, with 6 small rami. Palp rather robust (Figs. 9, 10): (numbers for male followed in parentheses by those for female). Femur 0.9 (0.95) and chela 1.15 X (1.35) as long as carapace. L/B of tro- chanter 1.5 (1.85), femur 2.6 (2.95), patella 2.8 (2.7), and chela (without pedicel) 2.7 (2.7); L/D of hand (without pedicel) 1.55 (1.45); movable finger L / hand L 1.15 (0.95). Chela sexually dimorphic; that of male more robust. with a conical protuberance on medial side of hand, and with movable finger distinctly bowed (Fig. 11); that of female more slender, without a protuberance on hand, and with mov- able finger only gently curved. Surfaces gran- ulate; most setae clavodentate. Trichobothria as shown in Fig. 11. Fixed finger with about 45 and movable finger with 45-50 cusped mar- ginal teeth, and 1-3 internal and 7-9 external accessory teeth. Venom apparatus developed only in movable finger. Legs more robust than those of T. aalbui: leg IV (Fig. 14) with L/D of femur + patella 3.8 and tibia 6.0. Leg I sex- ually dimorphic: that of male apparently rap- torial, with robust femur, patella and tibia, and elongate, curved tarsus (Fig. 12), L/D of fe- mur+patella 2.75 and tibia 3.6; leg I of female normal, slender (Fig. 13), L/D of fe- mur+patella 3.35 and tibia 4.55. Tarsus of leg IV with a short acuminate or denticulate tactile seta 0.75 length of segment from proximal end. Measurements. — Figures given first for holotype male, followed in parentheses by those for allotype female. Body L 3.55 (4.12). Carapace L 1.18 (1.20). Chelicera L 0.30 (0.355). Palp: trochanter 0.63 (0.605)/0.385 (0.325); femur 1.04 (1.10)/0.40 (0.37); patella MUCHMORE— NEW PSEUDOSCORPIONS IN CAVES 211 0.99 (1. 02)70.35 (0.38); chela (without pedi- cel) 1.36 (1.60)70.50 (0.59); hand (without pedicel) 0.69 (0.85)70.45 (0.585); pedicel L 0.13 (0.12); movable finger L 0.78 (0.82). Leg I: femur+patella L 0.895 (0.925); femur 0.38 (0.33)70.31 (0.215); patella 0.63 (0.525)70.30 (0.20); tibia 0.755 (0.59)70.21 (0.13); tarsus 0.47 (0.495)70.09 (0.08). Leg IV: fe- mur-fpatella 0.895 (0.925)70.235 (0.24); tibia 0.835 (0.835)70.14 (0.13); tarsus 0.58 (0.59)7 0.095 (0.095). Etymology. — The species is named for Darrell Ubick, who collected the type speci- mens. Remarks. — The first legs of the male look as though they might be very useful in seizing or holding prey, but there is no direct evidence that this is so. They might, rather, be used in grasping the female during courtship and sperm transfer, which, in some chemetid pseu- doscorpions, can involve rather complex ma- neuvers (see Weygoldt 1969). DISCUSSION Several other genera of chemetid pseudo- scorpions have medial protuberances on the palpal chela. Tuberochernes is easily distin- guished from Mirochernes Beier 1930 (from eastern U.S.), in which the male has a very large, distally directed, hooklike process (Hoff 1949: fig. 45C). And it differs from Inter- chernes Muchmore 1980 (from Baja Califor- nia, Mexico), where the process is a small, discrete, conical nubbin located at the base of the fixed finger and is present in both sexes (Muchmore 1980). Bituberochernes Much- more 1974 (from Florida and the West Indies), likewise, has a small process at base of the fixed finger, but it differs fundamentally from Tuberochernes in having a three-bladed che- liceral flagellum, distinctive female genitalia, and highly specialized setae on leg I of the male (see Muchmore 1974b, 1979). Petter- chernes Heurtault 1986 (from Brazil), with a large hump on the chelal hand, has a three- bladed flagellum, and broad, leaflike setae (Heurtault 1986). No other chemetid pseudo- scorpions are known to have protuberances on the chelal hand. Cordylochernes octentoctus (Balzan 1891) (from South Africa?) was orig- inally illustrated as having a triangular tuber- cle on the base of the fixed chelal finger (Bal- zan 1891: fig. 5); however, Vachon (1942), on reexamination of the unique type of the spe- cies, found that the protuberance was actually a bit of foreign material stuck to the surface of the finger. Of the genera mentioned above, Tubero- chernes is more closely related to Mirocher- nes and Interchernes, in the possession of a four-bladed flagellum, paired, long, slender, tubular spermathecae, and other characters (see Muchmore 1974a). In these characters also, it is close to Chernes Menge 1855, Di- nocheirus Chamberlin 1929 and Hespero- chernes Chamberlin 1924, all widely distrib- uted in the United States. In addition to the distinctive medial protu- berance on the chelal hand, males of Tubero- chernes have a uniquely modified leg I (more so in T. ubicki than in T. aalbui). All segments of the first legs are more robust than in fe- males and the tarsus is curved, so that, in T. ubicki especially, it appears useful for seizing or grasping. The exact nature of the modifi- cations of the anterior appendages is not known, but, as they are found only in the males, it might be supposed that they are somehow related to courtship and mating. On the other hand, known species of the genus are found only in caves, and these may be adaptations to some aspect of life in that hab- itat. Though it is common in cheliferid pseu- doscorpions, sexually dimorphic modification of the first legs is rare in chernetids. Repre- sentatives of only three chemetid genera have been known previously to be so modified, namely, Pachychernes Beier 1932 from South and Central America, Orochernes Beier 1968 from Nepal and Siberia and Bituberochernes Muchmore 1974 from Florida and the West Indies (see Muchmore 1996). In all of these, the modifications involve the occurrence of very long, or short, specialized, setae, which are not present in Tuberochernes species. Representatives of Tuberochernes are pres- ently known only from caves at moderately high elevations, T. aalbui in Poleta Cave, Westgard Pass, White-Inyo Mountains, Inyo County, California, at about 2200 m elevation, and r. ubicki in Fly Cave, Gardner Canyon, Santa Rita Mountains, Santa Cruz County, Ar- izona, at about 1600 m. The widely separated and restricted localities of T. aalbui and T. ubicki in California and Arizona strongly sug- gest that these species are relicts of a formerly widespread ancestral population, fragmented 212 THE JOURNAL OF ARACHNOLOGY by desertification in the intervening areas. Similar disjunct patterns of distribution in California and Arizona have been observed in several other groups of arachnids: the antro- diaetid spiders Aliatypus Janus Coyle 1974 and A. isolatus Coyle 1974 (see Coyle 1974); the vaejovid scorpions Uroctonites giulianii Williams & Savary 1991 and U. huachuca (Gertsch & Soleglad 1972) (see Williams & Savary 1991); the hubbardiid schizomids Hubbardia borregoensis (Briggs & Horn 1966) and H. wessoni (R.V. Chamberlin 1939) (see Reddell & Cokendolpher 1995); two spe- cies of the phalangodid harvestman genus Si- talcina Banks 1911 (Ubick & Briggs, un- publ.); and others. It will not be surprising if additional representatives of Tuberochernes are found in other montane or subterranean refugia in California and Arizona. ACKNOWLEDGMENTS I am greatly indebted to Rolf Aalbu, Der- ham Giuliani, and especially Darrell Ubick for providing the specimens upon which this study is based and for much valuable infor- mation about them. Many thanks are due to B.RM. Curcic, V.E Lee, and the editors for valuable comments on the manuscript. LITERATURE CITED Balzan, L. 1891. Voyage de M.E. Simon au Ven- ezuela (Decembre 1887-Avril 1888). Arachni- des. Chemetes (Pseudoscorpiones). Ann. Soc. Entomol. France, 60:497-552. Beier, M. 1932. Pseudoscorpionidea 11. Subord. C. Cheliferinea. Tierreich, 58:1-294. Beier, M. 1968. Ein neues Chernetiden-Genus (Pseudoscorp.) aus Nepal. Khumbu Himal, 3:17- 18. Coyle, F.A. 1974. Systematics of the trapdoor spi- der genus Aliatypus (Araneae: Antrodiaetidae). Psyche, 81:431-500. Harvey, M.S. 1991. Catalogue of the Pseudoscor- pionida. Manchester Univ. Press, Manchester, England. 726 pp. Harvey, M.S. 1992. The phylogeny and classifi- cation of the Pseudoscorpionida (Chelicerata: Arachnida). Invert. Taxon., 6:1373-1435. Heurtault, J. 1986. Petterchemes brasiliensis, genre et espece nouveaux de Pseudoscorpions du Bresil (Arachnides, Pseudoscorpionida, Chernetidae). Bull. Mus. Natn. Hist. Nat., Paris, (4) 8:351-355. Hoff, C.C. 1949. The pseudoscorpions of Illinois. Bull. Illinois Nat. Hist. Survey, 24:407-498. Menge, A. 1855. Ueber die Scheerenspinnen. Neueste Schrift. Naturforsch. Ges., Danzig, 5:1- 43. Muchmore, W.B. 1974a. Clarification of the genera Hesperochernes and Dinocheirus (Pseudoscor- pionida, Chernetidae). J. ArachnoL, 2:25-36. Muchmore, W.B. 1974b. Pseudoscorpions from Florida. 2. A new genus and species Bitubero- chernes mumae. (Chernetidae). Florida Entomol., 57:77-80. Muchmore, W.B. 1979. Pseudoscorpions from Florida and the Caribbean area. 8. A new species of Bituberochernes from the Virgin Islands (Chernetidae). Florida Entomol., 62:313-316. Muchmore, W.B. 1980. Interchernes, a new genus of pseudoscorpion from Baja California (Pseu- doscorpionida: Chernetidae). Southwest. Nat., 25:89-94. Muchmore, W.B. 1996. An unusual new Pachy- chernes from Panama and Mexico (Pseudoscor- pionida: Chernetidae). Entomol. News, 108:00- 00. Reddell, J.R. & J.C. Cokendolpher. 1995. Cata- logue, bibliography, and generic revision of the order Schizomida (Arachnida). Texas Mem. Mus., Speleol. Monogr., 4:1-170. Vachon, M. 1942. A propos du Cordylochernes oc- tentoctus Balzan (Pseudoscorpions). Bull. Mus. Natn. Hist. Nat., Paris, (2) 14:181-184. Weygoldt, P. 1969. The biology of pseudoscor- pions. Harvard Univ. Press, Cambridge. 145 pp. Williams, S.C. & WE. Savary. 1991. Uroctonites, a new genus of scorpion from western North America (Scorpiones: Vaejovidae). Pan-Pacific Entomol., 61:212-2^1. Manuscript received 14 July 1996, accepted 17 February 1997. 1997. The Journal of Arachnology 25:213-227 NEW SPECIES OF CHTHONIIDAE AND NEOBISIIDAE (ARACHNIDA, PSEUDOSCORPIONES) FROM MONTENEGRO, YUGOSLAVIA Bozidar P.M. Curcic, Rajko N. Dimitrijevic, and Slobodan E. Makarov: Institute of Zoology, Faculty of Biology, University of Belgrade, Studentski Trg 16, 11000 Belgrade, Yugoslavia ABSTRACT. The pseudoscorpions of the genera Chthonius C.L. Koch 1843 (Chthoniidae) and Roncus L. Koch 1873 (Neobisiidae) from Montenegro, Yugoslavia have been studied. Three new species, Chthon- ius {Chthonius) prove, Roncus hors, and R. davor are described. Diagnostic characters of the analyzed taxa are thoroughly described and figured. Taxonomic interrelationships and geographical distribution are briefly discussed. Including these three species, the family Chthoniidae occurs with five species in Mon- tenegro, the family Neobisiidae with 13 species. Only two cave (troglophilic) species of Chthonius C.L. Koch 1843 (subgenus Chthon- ius s. str.) (Chthoniidae) are presently known from Montenegro, Yugoslavia, viz. C. (C) ex- armatus Beier 1939 and C.(C.) porevid Cur- cic, Dimitrijevic & Makarov 1996; the former species inhabits a cave on Mt. Orjen, while the latter populates the Knezlaz Pecina Cave, Krivosije, Mt. Orjen, near Risan (Curcic et al. 1996b). To date, only two pseudoscorpions of the genus Roncus L. Koch 1873 (Neobisiidae) are known to inhabit Montenegro (Curcic et al. 1996a,b). These are: Roncus yaginumai Curcic, Curcic & Dimitrijevic 1996, from a cave on the isle of Vranjina, near Podgorica, and R. belbog Curcic, Dimitrijevic & Maka- rov 1996, from the Knezlaz Pecina Cave (also the type-locality of C (C) porevid) (Curcic et al. 1996a, b). The aim of this study is to present descrip- tions of three new species (one of Chthonius and two of Roncus), as well as to define their precise taxonomic status. With the new spe- cies described in the present study, the total number of the Chthoniidae inhabiting Mon- tenegro is now five, and of the Neobisiidae — 13 species (Curcic 1974, 1988). METHODS In the present study, material from three samples of pseudoscorpions collected in 1991 and 1992 has been examined. The first sample from a cave in the village Gomji Morinj, near Risan, Montenegro (Yugoslavia), contained two new taxa: Chthonius {Chthonius) prove new species, and Roncus hors new species. The other two samples from Mt. Durmitor (from the canyon of the Susica River and the village of Tepca, 1000-1100 m elev.), Mon- tenegro (Yugoslavia), contained another un- described species: Roncus davor new species. The new species described in this paper are probably endemic forms inhabiting either caves (C. prove new species and R. hors new species) or epigean habitats {R. davor new species). All studied pseudoscorpion speci- mens were mounted on slides in Swan’s fluid (gum chloral medium) and deposited in the collections of the Institute of Zoology, Faculty of Biology, University of Belgrade, Yugosla- via. All trichobothrial designations are in ac- cordance with Beier (1932). Terminology for pedipalpal and pedal podomeres follows Har- vey (1992). CHTHONIIDAE Daday 1888 Chthonius {Chthonius) prove new species Figs. 1-5; Table 1 Etymology.— In Slav mythology. Prove is the deity of justice (Petrovic 1995). Specimen examined. — -Holotype female, from a cave in the village Gomji Morinj, near Risan, Montenegro, Yugoslavia; 27 June 1991 (collected by I.M. Karaman, together with the holotype male of Roncus hors new species). 213 214 THE JOURNAL OF ARACHNOLOGY Description. — Carapace slightly longer than wider (almost quadrangular); epistome differ- entiated (Fig. 4). Neither eyes nor eye-spots de- veloped. Setal formula: m4m+6+4+2+4 = 22 setae (a single microseta in the preocular recess on either side). A pair of posterior and lateral setae of unknown size (broken). Carapace pale- yellowish and transparent. Tergites I-X and stemites IV-X smooth, en- tire and uniseriate. Tergal formula: 4-4-4-6-6- 6-6-6-6-6. Female genital area: stemite II with 8 setae clustered medially and posteriorly in the form of a triangle. Stemite III with 8 setae and 3 suprastigmatic microsetae along each stigma. Stemite IV with 7 posterior setae and 3 microsetae on either side. Stemites V-X each with 8-10 setae. Male genital area: un- known. Pleural membranes granulostriate. Cheliceral spinneret (galea) in the form of a small sclerotic tubercle (Fig. 5). Cheliceral palm with six setae and two or three accessory microsetae, movable hnger with one seta. Fixed cheliceral hnger with two distal large teeth and a row of eight pointed and contig- uous teeth which diminish in size proximally. Movable cheliceral hnger with a small isolat- ed tooth (just below the level of the galea), one large tooth, and a series of 12 triangular teeth, slightly asymmetrical and diminishing in size proximally. Dentition of cheliceral hn- gers as in Fig. 5. Galeal seta inserted basal to the teeth of the movable cheliceral hnger. Fla- gellum of 1 1 blades, one small blade proxi- mally and 10 blades twice this length, more or less in pairs, distally. The most distal mem- bers of the series are curved but all, to some extent, are pinnate on two sides. Manducatory process (apex of pedipalpal coxa) with two long setae, pedipalpal coxa with three setae. Trochanter short, other ped- ipalpal articles moderately elongate (Figs. 1, 2). Chelal hngers of almost equal size. Fixed chelal hnger is slightly S-shaped, and the movable hnger is somewhat curved inwards, or C-shaped (Fig. 2), Tip of hxed hnger (distal to et) bears 2 or 3 small distal teeth. Fixed chelal hnger with 29 triangular teeth which occupy almost the whole length of the hnger blade; distal and proximal members of this se- ries are close-set, whilst the median teeth are spaced and slightly asymmetrical. Movable chelal hnger with 16-18 teeth; distal and me- dian members are inclined backwards, and these are followed by small, low and asym- metrical teeth; at the level of b-sb, these teeth merge into a dental lamella. Chelal hngers are longer than chelal palm, and pedipalpal femur is slightly shorter than ehelal hngers, but al- most as long as carapace (Table 1). Trichobothriotaxy: ib and isb on chelal palm; hxed chelal hnger with a further six trichobothria {et, est, esb, eb, it, and ist, and a pair of accessory setae nearer to et than to the hnger tip); movable chelal hnger bears four trichobothria {t, st, sb, and b). Seta esb distal to eb, ist closer to esb than to eb and distal to the former; it close to esf, it-est at the level of t-sf, et close to accessory setae. Seta sb closer to b than to st, st nearer to t than to sb. Distance st-sb is almost 1.8X as long as b-sb\ distance t-st more than 8X as long as sb-st. Seta b at the level of ist (Fig. 2). Coxa II bears 7 or 8 spines, and coxa III has 3 or 4 spines which are elongate and hnely pinnate on two sides. Intercoxal tubercle with two small setae. Tibia IV, metatarsus IV, and tarsus IV each with a long tactile seta. The whole specimen is depigmented and delicate in appearance. Measurements and morphometric ratios are presented in Table 1. Distribution. — South Montenegro, Yugo- slavia, in a cave; probably endemic species. Diagnosis. — This new species is pheneti- cally most similar to C. (C.) ischnocheles re- ductus Beier 1939, from the Jama Pothole on the island of Giuppana (Sipun), Croatia, as well as to C. (C.) absoloni Beier 1939, from the Duzice Pecina near Trebinje, south Her- cegovina, Bosnia-Hercegovina. From C. (C) ischnocheles reductus, C. (C.) prove is easily distinguished by the color of the body (yellow, with reddish-brown appendages vs. pale, al- most transparent), in the presence/absence of eyes (present vs. absent), in the carapacal se- tation (18 vs. 22 setae), in the setal formula of tergites I-X (4-4-4-4-6-6-6-6-4-4 vs. 4-4-4- 6-6-6-6-6-6-6), in the pedipalpal chelal length to breadth ratio of females (5.00 vs. 4.055), in the number of teeth on the fixed chelal finger of females (48 vs. 29), in the pedipalpal chelal length of females (0.92 mm vs. 0.73 mm). The new species is clearly distinct from C. (C.) absoloni in a number of morphological traits: the carapacal setation (22 vs. 18 setae). CURCIC ET AL.— new pseudoscorpions from MONTENEGRO 215 Figures 1-5. — Chthonius (Chthonius) prove new species, holotype female. 1, Pedipalp (trichobothria omitted); 2, Pedipalpal chela (trichobothria omitted); 3, Leg IV; 4, Carapace; 5, Chelicera. Scale lines in mm. 216 THE JOURNAL OF ARACHNOLOGY Table 1. — Linear measurements (in mm) and selected morphometric ratios in Chthonius (Chthonius) prove new species, Roncus hors new species and Roncus davor new species, all from Montenegro, Yugoslavia. Abbreviations: TS = tactile seta, T = tritonymph, D = deutonymph. Character C. (C.) prove 9 R. hors S 99 R. davor 6S T D Body Length (1) 1.58 2.04 2.445-3.18 2.30-2.75 2.13 1.63 Cephalothorax Length (2) 0.51 0.60 0.64-0.73 0.48-0.69 0.55 0.40 Breadth 0.48 0.51 0.58-0.66 0.45-0.62 0.38 0.38 Abdomen Length 1.07 1.44 1.715-2.54 1.82-2.06 1.58 1.23 Breadth 0.64 0.69 0.96-1.31 0.86-0.99 0.75 0.58 Chelicerae Length (3) 0.46 0.35 0.40-0.48 0.40-0.425 0.34 0.25 Breadth (4) 0.27 0.25 0.22-0.24 0.23-0.24 0.18 0.12 Length of movable finger (5) 0.24 0.18 0.27-0.33 0.28 0.23 0.15 Length of galea 0.01 0.005 0.01 0.01 0.005 0.003 Pedipalps Length with coxa (6) 2.03 2.90 3.245-3.845 3.52-3.64 2.53 1.70 Length of coxa 0.31 0.48 0.55-0.61 0.51-0.55 0.425 0.25 Length of trochanter 0.23 0.36 0.38-0.47 0.44-0.45 0.32 0.22 Length of femur (7) 0.52 0.58 0.60-0.795 0.70-0.71 0.53 0.33 Breadth of femur (8) 0.12 0.18 0.205-0.25 0.20-0.22 0.16 0.13 Ratio 7/8 4.33 3.22 2.93-3.18 3.23-3.50 3.31 2.54 Ratio 7/2 1.02 0.97 0.94-1.09 1.03-1.46 0.96 0.825 Length of patella (tibia) (9) 0.24 0.48 0.555-0.64 0.57-0.62 0.41 0.27 Breadth of patella (tibia) (10) 0.13 0.22 0.26-0.33 0.27-0.28 0.195 0.14 Ratio 9/10 1.85 2.18 1.94-2.13 2.11-2.21 2.10 1.93 Length of chela (11) 0.73 1.00 1.16-1.33 1.30-1.31 0.845 0.63 Breadth of chela (12) 0.18 0.28 0.41-0.46 0.38-0.40 0.275 0.195 Ratio 11/12 4.055 3.57 2.83-2.89 3.275-3.42 3.07 3.06 Length of chelal palm (13) 0.27 0.46 0.52-0.64 0.59-0.63 0.40 0.31 Ratio 13/12 1.50 1.64 1.27-1.39 1.55-1.575 1.45 1.59 Length of chelal finger (14) 0.55 0.54 0.64-0.69 0.68-0.71 0.445 0.32 Ratio 14/13 2.04 1.17 1.08-1.23 1.08-1.20 1.11 1.03 Leg IV Total length 1.595 2.08 2.395-2.68 2.51 1.84 1.085 Length of coxa 0.22 0.36 0.40-0.47 0.38 0.34 0.20 Length of trochanter (15) 0.18 0.27 0.31-0.34 0.33 0.22 0.16 Breadth of trochanter (16) 0.12 0.12 0.13-0.17 0.17 0.10 0.09 Ratio 15/16 1.50 2.25 2.00-2.38 1.94 2.20 1.78 Length of femur + patella (17) 0.45 0.54 0.62-0.72 0.65 0.47 0.26 Breadth of femur + patella (18) 0.19 0.19 0.20-0.27 0.24 0.185 0.11 Ratio 17/18 2.37 2.84 2.67-3.10 2.71 2.54 2.36 Length of tibia (19) 0.30 0.46 0.535-0.59 0.58 0.40 0.205 Breadth of tibia (20) 0.085 0.10 0.11-0.12 0.13 0.10 0.075 Ratio 19/20 3.53 4.60 4.86-4.92 4.46 4.00 2.73 Length of metatarsus (21) 0.14 0.17 0.19-0.23 0.22 0.16 0.10 Breadth of metatarsus (22) 0.07 0.075 0.08 0.09 0.08 0.06 Ratio 21/22 2.00 2.27 2.375-2.875 2.44 2.00 1.67 Length of tarsus (23) 0.305 0.28 0.33-0.34 0.35 0.25 0.16 Breadth of tarsus (24) 0.04 0.06 0.075-0.08 0.08 0.07 0.06 CURCIC ET AL.— new pseudoscorpions from MONTENEGRO 217 Table 1. — Continued. Character C. (C) prove 9 R. hors 6 99 R. davor 66 T D Ratio 23/24 7.625 4.67 4.125-4.53 4.375 3.57 2.67 TS ratio — tibia IV 0.53 0.59 0.54-0.56 0.61 0.54 0.43 TS ratio — metatarsus IV 0.43 0.26 0.16-0.285 0.23 0.21 0.32 TS ratio — tarsus IV 0.26 0.31 0.32-0.35 0.37 0.33 0.36 the setation of tergites I-V (4-4~4-6"6 vs. 4-4- 4-4-6), in the pedipalpal chelal length to breadth ratio of females (4.055 vs. 5.80), in the number of spines on coxae II of females (7 or 8 vs. 5), in the form of both pedipalpal chelal palm and chelal finger (less elongate vs. more elongate) (Fig. 2; Beier 1939, fig. 3). Neobisiidae J.C. Chamberlin 1930 Roncus hors new species Figs. 6-12; Table 1 Etymology.— In Slav mythology. Hors is the God of Sun (Petrovic 1995). Specimen examined. — Holotype male, Figures 6-12. — Roncus hors new species, holotype male. 6, Carapace; 7, Epistome; 8, Chelicera; 9, Pedipalp; 10, Pedipalpal chela (trichobothria omitted); 11, Leg IV; 12, Genital area. Scales in mm. 218 THE JOURNAL OF ARACHNOLOGY from a cave in the village Gornji Morinj, near Risan, Montenegro, Yugoslavia; 27 June 1991 (collected by I.M. Karaman, together with the holotype female of C. (C.) prove new species). Description. — Epistome small (but dis- tinct), triangular and apically rounded (Figs. 6, 7). A single pair of eyes developed; eye lenses somewhat reduced and flattened. Setal formula: 4+7-f5+l+6 = 23 setae (male) (Fig. 6). Carapace reticulate throughout. Abdominal tergite setal formula (I-X): 6-9- 11-11-11-11-11-10-10-10. Both tergites I-X and sternites IV-X entire, uniseriate, and smooth. Twelfth abdominal segment with two pairs of small setae. Female genital area: un- known. Male genital area: sternite II with 12 long median and posterior setae (of these, 6 setae are retromarginal); sternite III with 5 (3 + 2) anterior, 10 posterior setae, and 3 su- prastigmatic setae on either side; sternite IV with 10 posterior setae and 3 microsetae along each stigma. Sternites V-X with 13-15-14-13- 13-13 setae. Galea distinct, low and rounded. Cheliceral palm with 6, movable finger with one seta (Fig. 8). Cheliceral dentition as in Fig. 8. Fla- gellum with one short proximal blade and sev- en longer blades distally, characteristic of the genus Roncus. Apex of pedipalpal coxa (manducatory pro- cess) with four long setae. Pedipalpal trochan- ter with a small tubercle. A small exterolateral tubercle on pedipalpal femur present; pedipal- pal femur and chelal palm with interior gran- ulations, patella (tibia) smooth (Fig. 9). A sin- gle tiny tubercle present on the interolateral side of the chelal palm. No group of micro- setae proximal to trichobothria eb and esb; in- stead, some small setae distal to eb and esb (6-8) present. Fixed chelal finger with 47 small, asymmetrical, and close-set teeth; mov- able finger with 47 small and contiguous teeth. Chelal fingers longer than chelal palm and only slightly shorter than pedipalpal fe- mur (Table 1). Trichobothrial pattern: ist slightly closer to est than to isb; sb equidistant from b and st\ st closer to t than to sb. Dis- tribution of trichobothria as illustrated in Fig. 10. Leg IV: tibia, metatarsus, and tarsus each with a long tactile seta. Morphometric ratios and linear measure- ments are presented in Table 1. Distribution. — South Montenegro, Yugo- slavia, in a cave; probably endemic species. Diagnosis. — This new species is easily dis- tinguished from its phenetically similar con- gener, R. yaginumai, by the setation of the car- apace (23 vs. 24-27 setae), by the form of the pedipalpal podomeres (stout vs. elongate) (Figs. 9, 10) (Curcic et al. 1996a), by the num- ber of teeth on the fixed (47 vs. 62-70) and movable chelal fingers (47 vs. 62 — 65), by the carapace length (0.60 mm vs. 0.81-1.02 mm), by the pedipalpal length (2.90 mm vs. 4.49- 5.33 mm), by the ratio of the pedipalpal femur length to breadth ratio (3.22 vs. 3.52 — 3.89), by the pedipalpal chelal length (1.00 mm vs. 1.64-1.94 mm), by the pedipalpal tibia length to breadth ratio (2.18 vs. 3.35-3.63), and by the body size (smaller vs. larger) (Table 1) (Curcic et al. 1996a). From another epigean species from Mon- tenegro R. davor new species, R. hors new species differs in many important respects: the form of the galea (lower vs. higher; Figs. 8, 29, and 30), in the cheliceral length of males (0.40-0.425 mm vs. 0.35 mm), in the pedi- palpal length of males (3.52-3.64 mm vs. 2.90 mm), in the shape of the pedipalpal chelal palm (almost globular vs. ovate), in the ped- ipalpal femur length of males (0.70-0.71 mm vs. 0.58 mm), in the pedipalpal chelal length of males (1.30-1.31 mm vs. 1.00 mm), in the walking leg IV length of males (2.51 mm vs. 2.08 mm), and in the body size (larger vs. smaller) (Table 1). Roncus davor new species Figs. 13-25; Table 1 Etymology. — In Slav mythology, Davor is a chthonic deity, the son of Triglav (Petrovic 1995). Specimens examined. — Holotype female, and allotype male, from the canyon of the Susica River, Mt. Durmitor (1100 m elev.), Montenegro, Yugoslavia, collected on 4 Au- gust 1992 by I.M. Karaman. Paratypes: 1$, 16,2 tritonymphs, and 1 deutonymph, from the village of Tepca, Mt. Durmitor (1000 m elev.), Montenegro, Yugoslavia, 5 August 1992, same collector (together with a speci- men of Neobisium sp.). Description (based on adults). — Epistome small and rounded, knob-like; (Fig. 14) or low and triangular (Fig. 20). A pair of small eyes (with flattened lenses) present (Figs. 13, 19). CURCIC ET AL.— new pseudoscorpions prom MONTENEGRO 219 Figures 13-18. — Roncus davor new species, holotype female. 13, Carapace; 14, Epistome; 15, Chelic- era; 16, Leg IV; 17, Right pedipalp (trichobothria omitted); 18, Pedipalpal chela (trichobothria omitted). Scales in mm. Setal formulae: 4+6 + 2+4+2+6 = 24 (fe- male) and 4+6 + 2+4+2+6 = 24 setae (male). Carapace reticulate throughout. Tergites I-X with 6-9-11-12-11-11-12-12- 11-9, 6-8-10-11-12-11-12-11-11-10 (females), and 6-8-10-11-10-11-11-10-9-9 setae (male). Abdominal tergites I-X and stemites V-X smooth, uniseriate, and entire. Female genital area: stemite II with 10-12 small setae, clus- tered into two groups on either side of the mid-line; stemite III with 10 or 11 posterior setae and 3 or 4 suprastigmatic setae on either side; stemite IV with 11 or 12 marginal setae and 3 small setae along each stigma. Male genital area (Fig. 22): stemite II with 14-17 median and posterior setae (of these, 9 or 10 are retromarginal); stemite III with 4-7 (2 + 2 or 3+4) anterior, 10—12 posterior setae, and 3 or 4 suprastigmatic setae on either side; ster- nite IV with 7-10 posterior setae and 3 mi- crosetae along each stigma. Stemites V-X with 14-13-15-15-14-12 and 14-15-13-14- 220 THE JOURNAL OF ARACHNOLOGY Figures 19-25. — Roncus davor new species, allotype male. 19, Carapace; 20, Epistome; 21, Pedipalpal chela (trichobothria omitted); 22, Genital area; 23, Pedipalp; 24, Cheliceral fingers; 26, Leg IV. Scale lines in mm. 15-14 (female) and 14-14-13-13-14-13 and 14-15-13-14-15-14 setae (male). Twelfth ab- dominal segment with two pairs of small se- tae. Cheliceral spinneret (galea) small, low, and rounded (Figs. 15, 24). Cheliceral palm with six setae, movable finger with one seta. Fla- gellum eight-bladed (1 short proximal blade and seven longer blades distally), character- istic of the genus Roncus. Apex of pedipalpal coxa with four long se- tae. Pedipalpal trochanter with a small tuber- cle, femur with a small exterolateral tubercle and interior granulations; patella (tibia) smooth; chelal palm either with interior (Fig. 17) or with both interior and exterior granu- lations (Fig. 23). Chelal palm ovate (dorsal view). No microsetae proximal to trichoboth- ria eb and esb\ instead, 5-8 microsetae distal to eb and esb present (Figs. 18, 21). Fixed chelal finger with (male) 53-56 and (female) 55-57 teeth, movable chelal finger with 54- 56 (male) and 55-57 teeth (female). Chelal fingers longer than chelal palm and distinctly shorter than pedipalpal femur (Table 1). Trich- obothrial pattern: ist equidistant from isb and est\ sb equidistant from b and st; st closer to t than to sb. Distribution of trichobothria as illustrated in Figs. 19, 21. Tibia IV, metatarsus IV and tarsus IV each with a long tactile seta (Fig. 25). Morphometric ratios and linear measure- ments are presented in Table 1. Distribution. — Montenegro, Yugoslavia; CURCIC ET AL.— new pseudoscorpions from MONTENEGRO 221 epigean (in high elevation leaf-litter, soil, and humus). Probably endemic to the area. Diagnosis. ““From R. yaginumai, this new species is easily distinguished by the form of the pedipalpal articles (more elongate vs. less elongate; Figs. 17, 23) (Curcic et al. 1996a), by the relative position of the trichobothrium ist (closer to est than to isb vs. equidistant from est and isb), by the pedipalpal length of females (4.49“5.33 mm vs. 3.245“3.845 mm), by the pedipalpal chelal length to breadth ratio of females (3.35“3.63 vs. 2.83“2.89), by the pedipalpal chelal length of females (L64“L69 mm vs. 1.16-4.33 mm). For comparison with R. hors new species see the ‘Diagnosis’ of that species. ACKNOWLEDGMENTS We acknowledge the help of LM. Karaman, who collected the specimens considered here- in. This study has been supported, in the form of travelling expenses, by the Serbian Minis- try of Science and Technology Grant 03E03, by the Serbian Academy of Sciences and Arts, and by the ‘Beobanka’ -Belgrade. LITERATURE CITED Beier, M. 1932. Pseudoscorpionidea. L Subord. Chthoniinea et Neobisiinea. In, Das Tierreich, 57:1“258; Berlin. Beier, M. 1939. Die Hohlenpseudoscorpione der BalkanhalbinseL Eine aes dem Material der ‘Biospeologica balcanica’ basierende Synopsis. Stud. Geb. allg. Karstforsch., Brunn, Biol. Sen, 4:1“83. Curcic, B.P.M. 1974. Arachnoidea. Pseudoscorpi- ones. Catalogus Faunae Jugoslaviae, Cons. Acad, Sci. Rei Publ. Soc. Foed. JugosL, Acad. Sci. Art. Slovenica, 3:1“36. Curcic, B.P.M. 1988. Cave-dwelling pseudo- scorpions of the Dinaric Karst. Acad. Sci. Art. Slovenica, IV: Hist. Nat., Opera, 26, Inst. Biol, loannis Hadzi, 8:1“192. Curcic, B.P.M., S.B. Curcic & R.N. Dimitrijevic. 1996a. Roncus yaginumai, a new pseudoscorpi- on from Montenegro, Yugoslavia (Pseudoscor- piones, Neobisiidae). Acta ArachnoL, 45:7-12. Curcic, B.P.M., R.N. Dimitrijevic & S.E. Makarov. 1996b. On two new pseudoscorpion species from Montenegro, Yugoslavia (Pseudoscorpi- ones, Arachnida). Rev. ArachnoL In press. Curcic, B.P.M., V. Lee & S.E. Makarov. 1993. New and little-known cavernicolous species of Chthoniidae and Neobisiidae (Pseudoscorpiones, Arachnida) from Serbia, Bijdr, Dierk., 62:167- 178. Harvey, M.S. 1992. The phylogeny and classifi- cation of the Pseudoscorpionida (Chelicerata: Arachnida). Invert. Taxon., 6:1373-1435. Petrovic, S. 1995. Mitologija -kultura -civilizacija. Cigoja stampa & Salus, Beograd, 1-579. Manuscript received 3 April 1996, revised 15 Feb- ruary 1997. 1997. The Journal of Arachnology 25:222-227 RESEARCH NOTE ASSESSMENT OF THE RANDOM AMPLIFIED POLYMORPHIC DNA (RAPD) TECHNIQUE FOR THE ANALYSIS OF SPIDER POPULATIONS The ability to determine the degree of ge- netic relatedness between different popula- tions(both geographic and morphologic) of spiders would be of great value in many ar- eas of spider biology. For example, it would allow the testing of the hypotheses that woodland fragments can act as habitat is- lands (Beaumont 1993) and that spiders can pass freely between real islands by aerial dispersal (Duffey 1956). It would also allow the investigation of the basis (genetic or en- vironmental) of the intra-species size and shape variation seen in geographically sep- arate populations of some species (e.g., Ly- cosidae: Trochosa terricola Thorell 1856 and Pardosa pullata (Clerck 1757) from widely separated peat-bog sites (Curtis & Stinglhammer 1986)). Multi-locus DNA profiling (“finger-print- ing” Jeffreys et al. 1985a,b, 1991) experi- ments using the human DNA probes 33.6 and 33.15 (Jeffreys et al. 1985a, b) on ge- nomic DNA of P. pullata had previously demonstrated that there were no sequences complementary to either of these probes in the genome of this lycosid (Beaumont 1993). For this reason, and also because spi- ders (particularly the smaller species) are so small that they might not contain sufficient DNA to generate a conventional DNA pro- file (5 pig; Bruford et al. 1992), a different approach to genetic analysis was adopted. The technique of random amplified poly- morphic DNA (RAPD) analysis (Williams et al. 1990) or DNA amplification fingerprint- ing (DAF -Gaetano- Anolles et al. 1991) has been successfully used in recent years to ex- amine genetic relationships in a wide variety of species, including plants (Virk et al. 1995), insects (Hadrys et al. 1993), humans and other mammals (Welsh et al. 1991; Wil- liams et al. 1990), and micro-organisms (Caetano-Anolles et al. 1991; Welsh & McClelland 1990). This technique uses an arbitrarily chosen oligonucleotide primer in a low stringency polymerase chain reaction (PCR — Saiki et al. 1988; Newton & Graham 1994) to generate a series of amplified prod- ucts of different sizes from a target (usually intact genomic) DNA. These products are then separated by conventional agarose or polyacrylamide gel electrophoresis (Sam- brook et al. 1990). The pattern of fragments produced is specific for the combination of primer and target DNA used (and the precise conditions of the PCR) and is attributed to the distribution of primer-complementary sequences within the target DNA. Since this is an amplification method, it can generate useful data from very small amounts of tar- get DNA (e.g., as little as 1 ng; Hedrick 1992). Also, because of the ease of synthesis of primers and the speed of the PCR itself, it is feasible to screen large numbers of primers under different experimental con- ditions until a combination producing a suit- able pattern of bands is found. A preliminary investigation was thus un- dertaken to assess the suitability of the RAPD method for the genetic analysis of spider populations. The study used six dif- ferent primers and genomic DNA isolated from siblings of two different broods of the theraphosid Brachypelma albopilosa (Val- erio 1980). This spider was chosen because it was possible to purchase broods of known parentage. It was hoped to identify one or more primer(s) that would reliably enable both identification of siblings and discrimi- nation between members of different broods. METHODS Two broods of juvenile (third instar) spec- imens of R. albopilosa of different parentage 222 HETTLE ET AL.^DNA ANALYSIS OF SPIDER POPULATIONS 223 ► A1 A2 A3 A4 A5 A6 A7 M 23130 9416 6557 4361 2322 2067 23130 9416 6557 4361 2 1 Figures 1, 2.— -Agarose gels (0.8%) showing intact, high molecular weight genomic DNA(^) iso- lated from specimens of Brachypelma albopilosa from two different broods: 1, brood A; 2, brood B. 1-7 are different individuals within each brood. Each sample represents 10% of the total amount of DNA isolated from each specimen, M = marker DNA: genomic DNA of bacteriophage X digested with the restriction enzyme TfmdIIL Sizes of the DNA fragments are indicated in base pairs (bp.). There is 0.5 pg of DNA in the 23,130 bp, band. The amount of DNA in each sample of Brachypelma albopilosa DNA was estimated by comparison with the bands in the marker lane as recommended by Bruford et al. (1992). were purchased from Ronald N. Baxter (En- tomological Supplies), 45 Chudleigh Crescent, Ilford, Essex, IG3 9AT, England, UK. These were designated A and B, maintained under starvation conditions for four days after deliv- ery (to ensure ablation of intestinal fauna) and then stored at “80 °C. Voucher material of these specimens has been deposited at the Paisley Natural History Museum, Paisley Mu- seum & Art Galleries, High Street, Paisley, PAl 2BA, Scotland, UK. DNA was extracted from these whole, frozen, individual spiders using a protocol based on a procedure for the isolation of intact, high molecular weight DNA from mammalian cells described by Sambrook et al. (1990) and described in full elsewhere (Beaumont 1993). After extraction, an aliquot of each DNA sample was run on an agarose gel to assess concentration, purity and integrity. Six oligonucleotide primers were synthe- sized for use in these experiments. Some of the sequences were chosen because of their usefulness in previous RAPD experiments (e.g., 1 & 4 by Caetano-Anolles et al. 1991); others were selected as they had been suc- cessfully used as probes in conventional DNA profiling experiments (2 & 3 by Weising et al. 1989; 2 by Debarro et al. 1994); still others were wholly arbitrary (5 & 6). The sequences of these primers (shown 5 '-3' from left to right) are as follows: primer 1: CGCGCCGG, primer 2: GATAGATAGATAGATA, primer 3: GACAGACAGACAGACA, primer 4: GTGATCGCAG, primer 5: GTAAAACGA- CGGCCAGT, primer 6: CTAGGTCTT- GAAAGGAGTGC. Polymerase chain reactions, using 16 ng of B. albopilosa genomic DNA as target, were performed as described (Welsh et al. 1991), except that the annealing tempera- tures were kept constant throughout all the 224 THE JOURNAL OF ARACHNOLOGY cycles of any one reaction. Two sets of re^ actions were carried out for each primer, the first with an annealing temperature of 40 °C, the second with an annealing temperature of 30 °C. Agarose and polyacrylamide gels were run as described in Tris-borate electro- phoresis (TBE) buffer (Sambrook et al. 1990). Agarose gels (0.8%) were used to as- sess the concentration, purity and integrity of intact genomic DNA; 5% polyacrylamide gels (± 7M urea) were used to resolve the products of the PCR reactions. If appropri- ate, gels were stained in a solution of ethid- ium bromide (0.5 |jLgml~’ in TBE) for 15- 30 min. Polyacrylamide gels were dried un- der vacuum at 80 °C (2 h), wrapped in plas- tic film (“Saran-Wrap”, Dow Chemical Co.) and any bands visualized by exposure of the gel to autoradiography film (‘'Hyperfilm- MP”, Amersham, UK) in an autoradiogra- phy cassette fitted with two intensifying screens for 1-4 days at —80 °C. RESULTS AND DISCUSSION Intact, high molecular weight genomic DNA was successfully and consistently iso- lated from B. albopilosa (Figs. 1, 2). The amount of DNA isolated from each spider was estimated to be 0.25-1.0 jjig, sufficient for many RAPD analyses to be performed on each individual. In other studies, intact, high molecular weight genomic DNA was successfully isolated from other spider spe- cies (Amaurobiidae: Amaurobius similis (Blackwall 1845); Linyphiidae: Lepthyphan- tes leprosus (Ohlert 1865); Lycosidae: Par- dosa pullata; Araeeidae: Zygiella x-notata (Clerck 1757)) by the same method (not shown). The genomic DNAs from two in- dividuals from each brood (3 & 4 from A, 6 & 7 from. B) were used in polymerase chain reactions with each of the six primers (separately). The reaction products were separated by polyacrylamide gel electropho- resis, the gels dried and the DNA fragments visualized by autoradiography. The results of these reactions are summarized in Tables 1-6, and examples of the banding patterns produced are shown in Figs. 3-6. It should be noted that the DNA fragments on several gels were also visualized by ethidium bro- mide staining before drying (not shown) and a comparison of the stained gel and autora- diograph showed that no additional bands 4 Figures 3, 4. — Autoradiographs showing the re- sults of RAPD analysis on individuals A3, A4, B6 and B7 using primer 3 with a PCR annealing tem- perature of: 3, 40°C; 4, 30 °C. were detected by autoradiography. Thus, it is unnecessary to use radio-labelled nucle- otides in the RAPD procedure, thus simpli- fying this process. The only primer that consistently failed to produce amplified products of discrete sizes was primer 3 (Figs. 3, 4). At both annealing temperatures, this primer generated many fragments of similar sizes that were not re- solved by the electrophoretic technique used, resulting in the appearance of long smears on the autoradiographs. This sug- gests that there are many binding sites com- plementary to this primer throughout the tar- HETTLE ET AL.— DNA ANALYSIS OF SPIDER POPULATIONS 225 6 Figures 5, 6. — Autoradiographs showing the results of RAPD analysis on individuals A3, A4, B6 and B7 using primer 4 with a PCR annealing temperature of: 5, 40 °C; 6, 30 °C. Products common to all four samples (>-), or to A3 and A4 only (►), or to B6 and B7 only (<1) are indicated. get DNA. It is possible that this primer is annealing to microsatellite loci and that an^ nealing is occurring at many overlapping sites. For all the other primers used, prod“ ucts of discrete sizes (different in both num- ber and size for each primer) were consis- tently produced. This suggests there are smaller numbers of complementary sites available to which these primers can anneal. The exact pattern produced by a primer var- Table 1. — Total number of DNA fragments generated by PCR amplification from each target DNA at each annealing temperature by primers 1, 2, 4-6. DNA Primer 1 Primer 2 Primer 4 Primer 5 Primer 6 30 °C U o U o 40 °C 30 °C 40 °C 30 °C U 0 O 30 °C 40 °C A3 2 5 5 2 15 11 5 9 8 7 A4 4 5 6 6 11 9 8 9 5 6 B6 8 5 7 8 15 11 6 13 5 6 B7 7 3 5 7 16 12 4 11 5 4 226 THE JOURNAL OF ARACHNOLOGY Table 2. — Number of common amplified DNA fragments generated from target DNAs A3 and A4 at each annealing temperature by primers 1, 2, 4-6. Annealing temperature Primer 30 °C U o o 1 2 3 2 5 0 4 9 9 5 5 9 6 2 6 Table 3. — Numbers of common amplified DNA fragments expressed as percentages of total number of DNA fragments generated from target DNAs A3 and (A4) at each annealing temperature by primers 1, 2, 4-6. Annealing temperature Primer 30 °C 40 °C 1 100 (50) 60 (60) 2 100 (83) 0 (0) 4 60 (82) 82 (100) 5 100 (63) 100 (100) 6 25 (40) 86 (100) ied according to the annealing temperature used during the PCR. Primer 4 provides good examples of the type of pattern pro- duced (Figs. 5, 6). At both annealing tem- peratures, this primer generated many am- plified products from each target DNA and the patterns of products observed were very similar from all four target DNAs. Due to these similarities, reliable differentiation be- tween inter- and intra-family relationships was not possible using this primer. And, in- Table 4. — Number of common amplified DNA fragments generated from target DNAs B6 and B7 at each annealing temperature by primers 1, 2, 4-6. Annealing temperature Primer 30 °C 40 °C 1 4 3 2 4 4 4 14 11 5 4 10 6 5 4 Table 5. — Numbers of common amplified DNA fragments expressed as percentages of total number of DNA fragments generated from target DNAs B6 and (B7) at each annealing temperature by primers 1, 2, 4-6. Annealing temperature Primer 30 °C 40 °C 1 50 (57) 60 (100) 2 57 (80) 50 (57) 4 93 (88) 100 (92) 5 67 (100) 77 (91) 6 100 (100) 67 (100) deed, for this same reason, none of the prim- ers used could reliably differentiate between these types of relationships (see Tables 1- 6). Although no primer was found that could allow both identification of siblings and dis- crimination between members of different broods, the feasibility of the procedure was established. Analysis of the organization and sequence of the B. albopilosa genome, particularly of any repetitive sequences that may be found, should permit the design of more useful primers. It should be noted, however, that there are known to be some problems associated with RAPD analysis. For example, and as in all DNA profiling analyses, it can be difficult to determine whether similar bands in different lanes are matching or not. Also, distinguishing het- erozygotes from homozygotes and establish- ing Mendelian patterns of inheritance may prove problematic. Finally, there can be problems of data reproducibility between experiments. Some of these problems, and methods that might be employed to over- Table 6.- — -Numbers of common amplified DNA fragments generated at each annealing temperature from target DNAs A3, A4, B6 and B7 by primers 1, 2, 4-6. Annealing temperature Primer 30 °C 40 °C 1 0 3 2 3 0 4 8 8 5 4 6 6 2 3 HETTLE ET AL.— DNA ANALYSIS OF SPIDER POPULATIONS 227 come them, are discussed by Williams et al. (1990) and Hedrick (1992). We are grateful to C.A. Hopkins, D.S.T. Ni- choll and J.R.M. Thacker for their comments on the manuscript. LITERATURE CITED Beaumont, D.J, 1993. Community ecology studies on spiders of island habitats in western Scotland. Ph.D. thesis. University of Paisley, Paisley, Scot- land. 237 pp. Bruford, M.W., O. Hanotte, J.EY. Brookfield & T. Burke. 1992. Single-locus and multilocus DNA fingerprinting. Pp. 225-269, In Molecular Genet- ic Analysis of Populations — A Practical Ap- proach (A.R. Hoelzel, ed.) IRL Press, Oxford. Caetano-Anolles, G., B.J. Bassam & P. Gresshoff. 1991. DNA amplification fingerprinting using very short arbitrary oligonucleotide primers. Biotechnology, 9:553-557. Curtis, D.J. & H.R.G. Stinglhammer. 1986. Wan- derers and webspinners on Scottish peatbogs: size variations. Actas X Congr. Int. Arachnol. Jaca/Espana, 1:219-223. Debarro, P, T. Sherratt, S. Wratten & N. MacLean. 1994. DNA fingerprinting of cereal aphids using (GATA) 4. European J. Entomol., 91:109-114. Duffey, E. 1956, Aerial dispersal in a known spider population. J. Anim. EcoL, 25:85-111. Hadrys, H., B. Schier water, S.L. Dellaporta, R. De- salle & L.W. Buss. 1993. Determination of pa- ternity in dragonflies by random amplified DNA fingerprinting. Molec. EcoL, 2:79-87. Hedrick, P 1992. Shooting the RAPDs. Nature, 355:679-680. Jeffreys, A.J., A. MacLeod, K. Tamaki, D.L. Neil & D.G. Monckton. 1991. Minisatellite repeat coding as a digital approach to DNA typing. Na- ture, 354:204—209. Jeffreys, A.J., V. Wilson & S.L. Thein. 1985a. Hy- pervariable “minisatellite” regions in human DNA. Nature, 314:67-73. Jeffreys, A.J., V. Wilson & S.L. Thein. 1985b. In- dividual specific “fingerprints” of human DNA. Nature, 315:76-79. Newton, C.R. & A. Graham. 1994. PCR — Poly- merase Chain Reaction. Bios Scientific Publ., Oxford. Saiki, R.K., D.H. Gelfand, S. Stoffel, S.J. Scharf, R. Higuchi, G.T Horn, K.B. Mullis & H.A. Er- lich. 1988. Primer-directed enzymatic amplifi- cation of DNA with athermostable DNA poly- merase. Science, 239:487-491. Sambrook, J., E.E Fritsch & T. Maniatis. 1990. Molecular Cloning — A Laboratory Manual (2nd ed.). Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Virk, P.S., B.V. Ford-Lloyd, M.T. Jackson & H.J. Newbury. 1995. Use of RAPD for the study of diversity within plant germplasm collections. He- redity, 74:170 — 179. Weising, K., E Weigand, A.J. Driesel, G. Kahl, H. Zischler & J.T. Epplen. 1989. Polymorphic sim- ple GATA/GACA repeats in plant genomes. Nucl. Acids Res., 17:10128. Welsh, J. & M. McClelland. 1990. Fingerprinting genomes using PCR with arbitrary primers. Nucl. Acids Res., 18:7213-7218. Welsh, J., C. Petersen & M. McClelland. 1991. Polymorphisms generated by arbitrarily primed PCR in the mouse: application to strain identifi- cation and genetic mapping. Nucl. Acids Res., 19:303-306. Williams, J.G.K., A.R. Kubelik, K.J. Livak, J.A. Rafalski & S.V. Tingey. 1990. DNA polymor- phisms amplified by arbitrary primers are useful as genetic markers. Nucl. Acids Res., 18:6531- 6535. Simon J.H. Hettle, Wendy Hall, Susan L. Dyas, David J. Beaumont, Dahlia M. Caplan & David J. Curtis: Dept, of Bio- logical Sciences, University of Paisley, Paisley PAl 2BE, Scotland, UK. Manuscript received 7 December 1995, revised 10 October 1996. 1997. The Journal of Arachnology 25:228 AAS FUND ANNOUNCEMENT The AAS Fund for Arachnological Re- search (AAS Fund) is funded and adminis- tered by the American Arachnological Soci- ety. The purpose of the fund is to provide re- search support for work relating to any aspect of the behavior, ecology, physiology, evolu- tion, and systematics of any of the arachnid groups. Awards may be used for field work, museum research (including travel), expend- able supplies, identification of specimens, and/ or preparation of figures and drawings for publication. Monies from the fund are not de- signed to augment or replace salary. Individual awards will not exceed $1000.00, and, although open to all students and faculty with less than $500 per year re- search budget, preference will be given to stu- dents. A total of $6000.00 is available for awarding during each funding year. A funding year has two granting periods: the Winter- Spring and Summer-Fall. Available monies could be expended for three large proposals during each granting period, or a greater num- ber of partially funded proposals, and/or a number of smaller, less expensive proposals. The final funding pattern is at the discretion of the review committee. Applications for support should be received by the chair of the review committee no later than May 30 or November 30, for funding by June 30 and December 30, respectively. To be considered for an award from the AAS Fund, please submit four copies of a proposal of no more than five pages (including references) detailing your research project. Proposals should have three main parts: 1) an INTRODUCTION where background in- formation is presented relative to the proposed work. The introduction should include a sec- tion which places the proposed work in con- text with currently known relevant informa- tion, a section which provides justification for the proposed work, and a clear statement of the hypothesis(ses) to be tested, or, in the case of systematic revisions, the type of synthesis that will be achieved and its significance; 2) a METHODS section where the methods, ma- terials, experimental design, and statistical or taxonomic analysis(ses) to be used are clearly and concisely presented, and 3) a BUDGET explaining (in detail) how monies awarded will be spent in the proposed research. Proposals should be submitted to: Dr. Elizabeth M. Jakob, AAS Fund Chair Department of Entomology 102B Femald Hall University of Massachusetts Amherst, MA 01003-2410 USA Proposals must be submitted in English. The four copies of the proposal must be in the hands of the Fund chair by the appropriate deadlines to be considered. Electronic sub- mission and/or FAX submission of proposals with hard copies to follow is acceptable only if the other copies arrive before the stated deadlines. If these submission rules are diffi- cult or prohibitive because of cost, erratic postal services, or remote location (remote field stations or sites), other methods of sub- mission may be acceptable. For other submis- sion possibilities, please contact the chair of the Fund at the above address, or electroni- cally at ejakob@bgnet.bgsu.edu. Alternative submissions will be accepted only if the chair has been previously contacted, and all dead- lines will still apply. 228 INSTRUCTIONS TO AUTHORS (revised October 1996) Manuscripts are preferred in English but may be ac- cepted in Spanish, French or Portuguese subject to availability of appropriate reviewers. Authors whose pri- mary language is not English may consult the Associate Editor for assistance in obtaining help with English manuscript preparation. All manuscripts should be pre- pared in general accordance with the current edition of the Council of Biological Editors Style Manual unless instructed otherwise below. Authors are advised to con- sult a recent issue of the Journal of Arachnology for additional points of style. Manuscripts longer than 1500 words should be prepared as Feature Articles, shorter papers as Research Notes. Send four identical copies of the typed material together with photocopies of illustra- tions to the Associate Editor. Do not send original handmade illustrations until the manuscript has been ac- cepted. Mail to: Petra Sierwald, Associate Editor; Division of In- sects, Dept, of Zoology, Field Museum, Roosevelt Road at Lakeshore Drive, Chicago, IL 60605 USA [Telephone: (312) 922-9410, ext. 841; FAX: (312) 663-5397; E-mail: SIERWALD@FMNH.ORG] Correspondence relating to the initial submission of a manuscript, as well as the review process, should be directed to the Associate Editor. Correspondence re- lating all other matters should be directed to the Editor. After the manuscript has been accepted, the author will be asked to submit the manuscript on a computer disc. It must be in a widely used program (preferably in MS DOS WordPerfect). Indicate clearly on the computer disc the word processing program used. FEATURE ARTICLES Title page. — The title page will include the com- plete name, address, and telephone number of the au- thor with whom proofs and correspondence should be exchanged, a FAX number and electronic mail address if available, the title in capital letters, and each author’s name and address, and the running head (see below). Abstract. — The heading in capital letters should be placed at the beginning of the first paragraph set off by a period. In articles written in English, a second abstract in an acceptable language may be included pertinent to the nationality of the author(s) or geo- graphic region(s) emphasized in the article. A second abstract in English must be included in articles not written in the latter language. Text. — Double-space text, tables, legends, etc. throughout. Three levels of heads are used. The first level (METHODS, RESULTS, etc.) is typed in capitals and on a separate line. The second level head begins a paragraph with an indent and is separated from the text by a period and a dash. The third level may or may not begin a paragraph but is italicized and sepa- rated from the text by a colon. Use only the metric system unless quoting text or referencing collection data. All decimal fractions are indicated by the period regardless of language of the text. Citation of references in the text: Cite only papers already published or in press. Include within parenthe- ses the surname of the author followed by the date of publication. A comma separates multiple citations by the same author(s) and a semicolon separates citations by different authors, e.g., (Smith 1970), (Jones 1988; Smith 1993), (Smith 1986, 1987; Smith & Jones 1989; Jones et al. 1990). Literature cited section. — Use the following style: Lombardi, S. J. & D. L. Kaplan. 1990. The amino acid composition of major ampullate gland silk (dragline) of Nephila clavipes (Araneae, Tetragnath- idae). J. Arachnol., 18:297-306. Krafft, B. 1982. The significance and complexity of communication in spiders. Pp. 15-66, In Spider Communications: Mechanisms and Ecological Sig- nificance. (P. N. Witt & J. S. Rovner, eds.). Princeton University Press, Princeton, New Jersey. Footnotes. — Footnotes are permitted only on the first printed page to indicate current address or other information concerning the author. All footnotes are placed together on a separate manuscript page. Running head. — The author surname(s) and an ab- breviated title should be typed all in capital letters and must not exceed 60 characters and spaces. The running head should be placed near the top of the title page. Taxonomic articles. — Consult a recent taxonomic article in the Journal of Arachnology for style or con- tact the Editor. Tables. — Each table, with the legend above, should be placed on a separate manuscript page. Only hori- zontal lines (usually three) should be included. Use no footnotes; instead, include all information in the leg- end. Make notations in the text margins to indicate the preferred location of tables in the printed text. Illustrations. — Address all questions concerning il- lustrations to: James W. Berry, Editor; Dept, of Biological Sci- ences, Butler University, Indianapolis, Indiana 46208 USA [Telephone (317) 940-9344; FAX (317) 940-9519; E-mail: BERRY@BUTLER.EDU] All art work must be camera-ready for reproduction. In line drawings, pay particular attention to width of lines and size of lettering when reductions are to be made by the printer. Multiple photos assembled on a single plate should be mounted with only a minimum of space separating them. In the case of multiple illus- trations mounted together, each illustration must be numbered sequentially rather than given an alphabetic sequence. Written on the back should be the name(s) of author(s) and an indication of top edge. The author should indicate whether the illustrations should be one column or two columns in width. The overall dimen- sions of original artwork should be no more than 1 1 inches (28 cm) X 14 inches (36 cm). Photocopies in review manuscripts should be reduced to the exact measurements that the author wants to appear in the final publication. Larger drawings present greater dif- ficulty in shipping and greater risks of damage for which the JOA assumes no responsibility. Make no- tations in the text margins to indicate the preferred position of illustrations in the printed text. Legends for illustrations should be placed together on the same page(s) and separate from the illustrations. Each plate must have only one legend, as indicated below: Figures 1-4. — A-us x-us, male from Timbuktu. 1, Left leg; 2, Right chelicera; 3, Dorsal aspect of geni- talia; 4, Ventral aspect of abdomen. Figures 27-34. — Right chelicerae of species of A-us from Timbuktu. 27, 29, 31, 33, Dorsal views; 28, 30, 32, 34, Prolateral views of moveable finger; 27, 28, A- us x-us, holotype male; 33, 34, A-us y-us, male. Scale = 1.0 mm. Assemble manuscript for mailing. — Assemble the separate sections or pages in the following sequence: title page, abstract, text, figure legends, footnotes, ta- bles with legends, figures. Page charges, proofs and reprints. — There are no page charges, but authors will be charged for changes made in the proof pages. Authors will receive a reprint order form along with their page proofs. Reprints will be billed at the printer’s current schedule of costs. RESEARCH NOTES Instructions above pertaining to feature articles ap- ply also to research notes, except that abstracts and most headings are not used and the author’s name and address follow the Literature Cited section. CONTENTS The Journal of Arachnology Volume 25 Feature Articles Number 2 Salticidae of the Pacific Islands. II. Distribution of Nine Genera, with Descriptions of Eleven New Species by James W. Berry, Joseph A. Beatty and Jerzy Proszyhski 109 Theraphosidae of the Mojave Desert West and North of the Colorado River (Araneae, Mygalomorphae, Theraphosidae) by Thomas R. Prentice 137 Callobius guachama (Araneae, Amaurobiidae): Habitat, Distribution and Description of the Female by Richard S. Vetter and Thomas R. Prentice 177 Foraging Versatility and the Influence of Host Availability in Argyrodes trigonum (Araneae, Theridiidae) by Karen R. Cangialosi 182 Bionomics of the Spider, Crossopriza lyoni (Araneae, Pholcidae), a Predator of Dengue Vectors in Thailand by Daniel Strickman, Ratana Sithiprasasna and Dawn Southard 194 Leg Autotomy and Avoidance Behavior in Response to a Predator in the Wolf Spider, Schizocosa avida (Araneae, Lycosidae) by Fred Punzo 202 Tuberochernes (Pseudoscorpionida, Chernetidae), a New Genus with Species in Caves in California and Arizona by William B. Muchmore 206 New Species of Chthoniidae and Neobisiidae (Arachnida, Pseudoscorpiones) from Montenegro, Yugoslavia by Bozidar P.M. Curcic, Rajko N. Dimitrijevic and Slobodan E. Makarov 213 Research Notes Assessment of the Random Amplified Polymorphic DNA (RAPD) Technique for the Analysis of Spider Populations by Simon J.H. Hettle, Wendy Hall, Susan L. Dyas, David J. Beaumont, Dahlia M. Caplan & David J. Curtis 222 Announcement Arachnological Research Fund 228 The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 25 1997 NUMBER 3 THE JOURNAL OF ARACHNOLOGY EDITOR: James W. Berry, Butler University ASSOCIATE EDITOR: Petra Sierwald, Field Museum EDITORIAL BOARD: A. Cady, Miami (Ohio) Univ. at Middletown; J. E. Carrel, Univ. Missouri; J. A. Coddington, National Mus. Natural Hist.; J. C. Cokendolpher, Lubbock, Texas; F. A. Coyle, Western Carolina Univ.; C. D. Dondale, Agriculture Canada; W. G. Eberhard, Univ. Costa Rica; M. E. Galia- no, Mus. Argentino de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia, Missouri; C. Griswold, Calif. Acad. Sci.; N. V. Horner, Midwestern State Univ.; D. T. Jennings, Garland, Maine; V. F. Lee, California Acad. Sci.; H. W. Levi, Harvard Univ.; E. A. Maury, Mus. Argentino de Ciencias Naturales; N. I. Plat- nick, American Mus. Natural Hist.; G. A. Polis, Vanderbilt Univ.; S. E. Riechert, Univ. Tennessee; A. L. Rypstra, Miami Univ., Ohio; M. H. Robinson, U.S. National Zool. Park; W. A. Shear, Hampden-Sydney Coll.; G. W. Uetz, Univ. Cincinnati; C. E. Valerio, Univ. Costa Rica. The Journal of Arachnology (ISSN 0160-8202), a publication devoted to the study of Arachnida, is published three times each year by The American Arach- nological Society. Memberships (yearly): Membership is open to all those in- terested in Arachnida. Subscriptions to The Journal of Arachnology and American Arachnology (the newsletter), and annual meeting notices, are included with mem- bership in the Society. Regular, $30; Students, $20; Institutional, $80 (USA) or $90 (all other countries). Inquiries should be directed to the Membership Secretary (see below). Back Issues: Patricia Miller, P.O. Box 5354, Northwest Mississippi Community College, Senatobia, Mississippi 38668 USA. Telephone: (601) 562- 3382. Undelivered Issues: Allen Press, Inc., 1041 New Hampshire Street, P.O. Box 368, Lawrence, Kansas 66044 USA. THE AMERICAN ARACHNOLOGICAL SOCIETY PRESIDENT: Ann L. Rypstra (1997-1999), Dept, of Zoology, Miami Univer- sity, Hamilton, Ohio 45011 USA. PRESIDENT-ELECT: Frederick A. Coyle (1997-1999), Department of Biology, Western Carolina University, Cullowhee, North Carolina 28723 USA MEMBERSHIP SECRETARY: Norman I. Platnick (appointed), American Museum of Natural History, Central Park West at 79th St., New York, New York 10024 USA. TREASURER: Gail E. Stratton, Department of Biology, University of Missis- sippi, University, Mississippi 38677 USA. BUSINESS MANAGER: Robert Suter, Dept, of Biology, Vassar College, Pough- keepsie, New York 12601 USA. SECRETARY: Alan Cady, Dept, of Zoology, Miami Univ., Middletown, Ohio 45042 USA. ARCHIVIST: Vincent D. Roth, Box 136, Portal, Arizona 85632 USA. DIRECTORS: H. Don Cameron (1997-1999), Matthew Greenstone (1997- 1999), Robert Suter (1995-1997). HONORARY MEMBERS: C. D. Dondale, W. J. Gertsch, H. W. Levi, A. F. Millidge, W. Whitcomb. Cover: Female Araneus pima wrapping prey. Taken with Nikon FM2 and 105mm Macro lens with manual flash, Fuji velvia film. Photo by Bryan Reynolds of Albuquerque, New Mexico. Publication date: 29 December 1997 @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). 1997. The Journal of Arachnology 25:229-244 RE-DESCRIPTION OF TOGWOTEEUS BICEPS (ARACHNIDA, OPILIONES, SCLEROSOMATIDAE) WITH NOTES ON ITS MORPHOLOGY, KARYOLOGY AND PHENOLOGY Robert G. Holmberg^: Centre for Natural and Human Science, Athabasca University, Athabasca, Alberta T9S 3A3 Canada James C. Cokendolpher: 2007 29th Street, Lubbock, Texas 79411 USA ABSTRACT. The harvestman genus Togwoteeus Roewer 1952 is monotypic. Its only species, T. biceps (Thorell 1877), is known from throughout western Canada and USA and is newly recorded from Cali- fornia, Nebraska, Nevada, Oregon, South Dakota, and Utah. This species occurs below 500 m and at the highest elevation (4100 m) of any recorded harvestman in North America. It ranges from about 33-54° N and exhibits considerable variation in its morphological measurements. Twenty-seven such measure- ments are summarized for 80 males and 74 females. Ultrastructural details of the integument, appendages and genital organs are presented. The karyotype is 2n = 22 for both sexes. All chromosomes are meta- centric and obvious sex chromosomes were not detected. Immature T. biceps overwinter and reach adult- hood in late spring or early summer. Adults die by late fall. Members of the monotypic genus Togwo- teeus Roewer 1952 are common harvestmen of prairie and mountain habitats in western regions of Canada and USA. For nearly 100 years, the single species currently recognized was known by the name Homolophus biceps (Thorell 1877). Cokendolpher (1987) pro- posed the new combination Togwoteeus bi- ceps because true members of the genus Hom- olophus Banks 1894 are known only from Asia. Togwoteeus was transferred from the family Phalangiidae to the Sclerosomatidae by Crawford (1992). Examination of museum specimens revealed considerable variation within this species, e.g., specimens from Brit- ish Columbia and Idaho had exceptionally longer legs. Because this species is found over such a large region of North America and was poorly studied, we sought to determine if the observed morphological variations were due to the existence of unrecognized new species. While doing this study we documented mor- phological, chromosomal, and phonological information which we herein present. METHODS Collections from which samples were ex- amined, other than those of the authors, are To whom correspondence should be addressed. listed in the Acknowledgments. Specimens used for measurements are described in Table 1. Those from Edmonton and Writing-on- Stone are retained by RGH. Those from Wy- oming and Utah are in the American Museum of Natural History. One count and 27 mea- surements were made: four of the body, eight of the pedipalp, femora I-IV, tibiae I-IV, three of the genital operculum, four of the ocular tubercle; and the number of metatarsal bands on leg II. Measurements, examination meth- ods, and terminology were as presented by Cokendolpher (1981), except the palpal femur and tibia lengths were measured along their greatest lengths (i.e., in lateral view, longitu- dinal distance between diagonal lines drawn from basoventral point to proximodistal point. Fig. 36, fl and tl). The ovipositors were cleared with lactophenol for examination and drawing of the seminal receptacles. Statistics were analyzed with the statistical computer program SPSS^. For morphological mea- surements two-tailed f-tests were used. For the single count, the Kolmogorov- Smirnov two- sample test was used. Alcohol preserved spec- imens were critical point dried, sputter coated with gold and photographed with scanning electron microscopes (SEM). Photographs 229 230 THE JOURNAL OF ARACHNOLOGY Table 1. — Specimens used for body measurements. Location Latitude N, Longitude W Altitude (m) Collection dates Males Females Edmonton, Alberta 53°46\ 113°25' 580 May- August 1982 20 20 Writing-on-Stone Provincial Park, Alberta 49°05', lir38' 915 May- July 1981 20 20 Grand Teton National Park, Wyoming 43°29', 110°50' 2260-2440 July-August 1962, August 1969 20 14 Timpanogos, Utah County & Mill Creek Canyon, Salt Lake County, Utah 40°26-4r, lir30-50' 1585-1725 May-June 1931-1941 20 20 were also taken through dissecting micro- scopes. Methods for the preparation of the karyotypes are reported by Cokendolpher & Brown (1985). The karyotypes were done in July 1983 from gonads of sUbadult specimens collected at the Logan Canyon summit. Cache County, Utah. The specimens were collected under rocks in the snow and transported alive in an ice-chest to Lubbock, Texas. They were maintained in the refrigerator until dissected. RESULTS AND DISCUSSION Togwoteeus Roewer Mitopus: Thorell 1877:525; Banks 1893:206. Homolophus: Banks 1894:160, 163, 164 (in part); 1900:123; 1901:674; Cockerell 1911:253; Roewer 1910:259 (in part); 1912:31 (in part); 1923:879-880 (in part); 1929:2 (in part); 1952: 268 (in part); 1957:355 (in part); 1960:24, 30 (in part); Comstock 1912:66, 71; 1940:66, 71; Kastner 1937:392 (in part); Katayama & Post 1974:13-14; Cokendolpher & Cokendolpher 1982:1215; Cokendolpher 1985:371, 399 (in part); 1987:89, 94 (in part). Togwoteeus Roewer 1952:268; 1957:356; Crawford 1992:45; Cokendolpher & Lee 1993:16. Type species. — Togwoteeus granipalpus Roewer 1952; by monotypy. Junior subjective synonym of Mitopus biceps Thorell 1877 and Homolophus punctatus Banks 1894. Diagnosis. — Body with thick, hard, tuber- culate-microgranulate cuticle; off-center mi- cropores present on dorsal tubercles; preocular area without mound but with two groups of small denticles near anterior margin edge; su- pracheliceral lamellae well developed and toothed; pedipalps without apophyses on dis- tal ends of patellae or tibiae in juveniles or adults, without campaniform organ on distal end of femora, claw smooth, not toothed; male pedipalps enlarged, tarsus bulbous at base and with ventral teeth; legs generally short, femur I usually equal to or shorter than body length, no pseudoarticulary nodules in femora; leg coxae without lateral rows of denticles; penis alate, i.e., with wing-like extensions. Comparisons. — The presence of an alate penis and a smooth palpal claw separate Tog- woteeus from all sclerosomatid opilions, ex- cept Leuronychus Banks 1900 and Cosmobu- nus Simon 1879. These genera can be easily distinguished from Togwoteeus by their longer legs (femur I always much longer than the body length) and lack of denticles in front of the ocular tubercle. Subordinate taxa. — The genus is mono- typic. Distribution. — Western North America (Fig. 1). Togwoteeus biceps (Thorell) (Figs. 1-38) Mitopus biceps Thorell 1877:525-528; Pavesi 1889:531; Banks 1893:206, 207; Cokendolpher 1987:94; Crawford 1992:45. Homolophus biceps: Banks 1894:163; 1895:431; 1900:123; 1901:674; 1902:593; 1916:72; Cock- erell 1907:620; 1911:253; Roewer 1912:31; 1923:880; 1957:355; Comstock 1912:71; 1940: 71; Levi & Levi 1951:219, 221, fig. 1; 1955:32; 1968:245; Goodnight & Goodnight 1953:175; Holmberg 1970:127-129, figs. 3. 4, 3. 7, A. 1; Schmoller 1970:127-128, 132; 1971a:323, 327; 1971b:346, 348; Bragg & Leech 1972:67; Katay- ama & Post 1974:8-10, 13, 14, 20, fig. 1; Holm- berg et al. 1981:19; Holmberg & Kokko 1983: 49-52, figs. 1-4; Cokendolpher 1985:399; 1987: 94; Poinar 1985:122. Togwoteeus biceps: Cokendolpher 1987:94; Cok- endolpher 1993:129, 132, 138; Cokendolpher & Lee 1993:16, 25-31. HOLMBERG & COKENDOLPHER— RE-DESCRIPTION OF TOGWOTEEUS BICEPS 231 Figure 1. — Distribution of Togwoteeus biceps in western North America. Collection sites that had nearly the same location were not mapped. Togwoteeus (= Homolophus) biceps: Edgar 1990: 567. Homolophus punctatus Banks 1894:164; 1901:674; Roewer 1923:880-881; Cokendolpher 1987:94. Togwoteeus granipalpus Roewer 1952:268, fig. 2; 1957:356; Levi & Levi 1955:32; Cokendolpher 1985:399; 1987:94; Crawford 1992:45. Globipes sp.: Blake 1945:232 (misidentification). Types. — The holotype of Togwoteeus granipalpus is labeled “Dr. E.C. Zimmerman 18.8.1951 Opiliones. Leptobuninae No. 14 Togwoteeus granipalpus Rwr IM Genotypus n.g. n.sp. Wyoming. Togwotee Pass 2743 m, Teton Co., Zimmerman Leh Rwr det. 1952” and is housed in Senckenberg Natur-Museum (RII/ 11047/14). JCC examined the specimen and found it to be a female. The types of Mitopus biceps were from USA: Idaho, no specific locality (5 July), and Colorado, [Clear Creek County] Gray’s Peak (a little below the summit), (7 July). These were apparently lost by the early part of the 20th century (Roewer 1923). Dr. Arbocco (pers. comm. 1981) checked the collection at Museo Civico di Storia Naturale “Giacoma Doria” twice for the types of this species without success and agreed that the lectotype should be designated from material collected by Thorell at the Naturhistoriska Riksmuseet. Thus, we are designating the male and female from the Naturhistoriska Riksmuseet as the lectotype and paralectotype, respectively. They have both been labeled as such: “det. Cokendolpher 1981”. The original data labels in the vial with the types correspond with those published except for the date: “Riks- museets Entomologiska Afdelnig. Collectio T. Thorell Mitopus biceps Thor. Idaho U.S.A. (Packard) no. 73”, “March 2 1891”, and “Mitopus biceps Thor. Idaho U.S.A. Packard Md.” As the labels are of unknown authorship it is possible they are incorrect. The speci- mens closely match ThorelTs (1877) detailed written description. The types of Homolophus punctatus were from the USA: Olympia, Thurston County, Washington (Id, Trevor Kincaid) and Bear, Adams County, Idaho (1$, L.M. Cockerell). These have apparently been lost. They were not examined by Rower (1923). Distribution. — Three most western prov- inces of Canada and 14 western states of USA (Fig. 1). The records for California, Nebraska, Nevada, Oregon, South Dakota, and Utah are the first published for these states. The female (SMF RII/2663/5) reported by Roewer (1957) from Pueblo, Mexico, is either misidentified or mislabeled. The specimen cannot be located at Senckenberg Natur-Mu- seum (Grasshoff pers. comm. 1981), and pos- sibly originated from Pueblo, New Mexico, or the well-known Pueblo, Colorado. We are unable to locate the specimens re- ported from near Centennial, Wyoming by Blake (1945) as Globipes sp. They are cer- tainly misidentified, as no known Globipes sp. occurs in or near Wyoming, the nearest re- cords being from Arizona and New Mexico (JCC pers. obs.). Blake’s specimens are prob- ably members of Togwoteeus, as this is the only genus of the region which resembles Globipes. We have also examined a series of T. biceps collected in Medicine Bow Moun- tains, near Blake’s collection site. Records. — (Based only on specimens examined; including adults and some immatures.) CANADA. Alberta: Athabasca; 30 km N of Athabasca; Big Hill Springs Provincial Park, near Cochrane; Etzi- komi; Lodgepole Pine Campground area. Cypress Hills Provincial Park; Edmonton; Elkwater; Hwy. 48, N Elkwater Provincial Park; Etzikom; Leth- 232 THE JOURNAL OF ARACHNOLOGY bridge; 16 km S Magrath; Medicine Hat; Seven Persons; Canadian Forces Base, Suffield; Waterton Lakes; Chief Mountain Road, Waterton Lakes Na- tional Park; Lewis Overthrust, Waterton Lakes Na- tional Park; Lookout Butte, Waterton Lakes Nation- al Park; Writing-on-Stone Provincial Park. British Columbia: Anarchist Mountain; Apex Mountain, near Keremeos; near Inkaneep Park along Okanog- an River; Kamloops; Kleena Kleene; Manning Pro- vincial Park; Beaver Pond Trail, Manning Provin- cial Park; isolated small park 6.4 km N Oliver; Meyers Flats, Oliver; Vaseaux Lake, Oliver; Eco- logical Reserve, Osoyoos; Kruger Springs, Oso- yoos; Osoyoos Lake; Salmon Arm; Summerland; Rattlesnake Point, Vernon; White Lake, 10 km S of Penticton. Saskatchewan: Buffalo Pound Provincial Park; Central Block and West Block, Cypress Hills Provincial Park; 17.7 km S Cypress Hills Provincial Park; Cypress Lake; 9.7 km WSW Dundum; Fort Walsh; Killdeer badlands; Saskatoon; Beaver Creek Park, 12.9 km SSW Saskatoon; 4.8 km NE Saska- toon; 8 km S Saskatoon; 20.9 km N Saskatoon; 22.5 km SSW Saskatoon; Saskatchewan Landing; Saskatchewan Landing Provincial Park. USA. Ari- zona: Coconino County: Kaibab National Forest, Grand Canyon. California: Lassan County: Blue Lake. Colorado: County?: Mummy Range (3353 m); Pingree Park. Archuleta County: Pagosa Springs. Boulder County: Arapaho Pass; Boulder Canon (2246 m); Longs Peak Valley, Rocky Moun- tain National Park (2774 m); Longs Peak, Rocky Mountain National Park (3400 and 4100 m). Chaf- fee County: Upper Spring Creek, near Monarch Pass. Clear Creek County: Loveland Pass; Mt. Evans; Summit Lake, Mt. Evans (3900 m). Conejos County: at top of Cumbres Pass; Trujillo Meadow Camp, 4.8 km N Cumbres (3048 m). Custer Coun- ty: West Cliff (= Westcliffe). El Paso County: on mountain side above Hwy 70, Pikes Peak; Pikes Peak (3048, 3658, 4115m); Canyon, Pikes Peak; Printing Office, Pikes Peak (3048 m). Freemont County: Wet Mountains, Stations 40, 47, 50 and 52 (2316, 2164, 2072, and 2127 m). Gilpin County: Gilpin. Grand County: Arapaho Peak (3962 m); Milner Pass. Gunnison County: Meyers Gulch. La Plata County: Cascade; Eldora (2438 m); Electra Lake (2560 m); Ward (2743 m). Larimer County: Glen Haven; near Long’s Park campground south of Estes Park Village; 16.1 km W Estes Park, Rocky Mountain National Park; Big Thompson Canyon, 1 1.3 km S Estes Park; Rustic. Pitkin Coun- ty: Aspen (3292 m). Rio Blanco County: small run beside route 13 ca. 9.7 km S Axial. Saguache County: Cochetopa Pass; Rist Canon, Fort Collins. Summit County: Hoosier Pass, south of Brekenrid- ge. Teller County: near Florissant. Idaho: Adams County: Summit, 1 1.3 km NE Council; canyon east of Meadows; north end of Payette Lake. Bear Lake County: Brentwood Lodge, Fish Haven; George- town; Montpelier. Boise County: 12.9 km S Galena (= Gardena?) Summit, junction Cherry and Coyote Creeks; Boise River, near entrance of north fork; Boundary Creek, Boise National Forest. Bonner County: Hana Flats, 8 km SW Nordman. Bonne- ville County: 16.1 km S Swan Valley. Cassia Coun- ty: Rock Creek Canyon, 24 km S of Rock Creek; Sublett Reservoir. Clark County: Spencer. Custer County: Salmon River Gorge, above Challis. Fre- mont County: St. Anthony. Kootenai County: Coeur d’Alene. Lincoln County: Little Wood River, Pagari. Nez Perce County: 8 km NW Culdesac. Twin Falls County: no specific location. Washing- ton County: 11.3 km NE Cambridge. Montana: Carbon County: E Rosebud (1890 m); Medicine Lake. Gallatin County: 9.7 km W Belgrade, W Gallstin River. Glacier County: 7 km W Browning. Granite County: Clark Fork near Bearmouth; Nim- rod. Ravalli County: Gird’s Creek, Hamilton. Sand- ers County: Thompson Falls. Nebraska: Dawes County: Belmont. Nevada: Elko County: Thomas Canyon Camp, 14.5 km SSE Lamoille Ruby Moun- tains (2286 m); Ruby Valley. Lander County: Kingston Camp, 48,3 km S Austin, Toiyabe Range. Washoe County: Reno. New Mexico: Bemialillo County: Sandia Mountains; near Crest, Sandia Mountains. Los Alamos County: Camp May (2900 m). Otero County: Bluff Springs, 14.5 km S Cloud- croft; Camp Deerhead, 1.6 km S Cloudcroft; Lin- coln National Forest, Fir Forest Campground. San Miguel County: Gallinas Canyon, NW of Las Ve- gas; just W Cowles; Lake Kathrine Trail, Cowles; Penitente Park, Cowles; Spirit Lake Trail, Cowles. Sandoval County: Jemez Mountains; Sandia Moun- tains. Santa Fe County: Lake Peak NE of Santa Fe; near ski area NE Santa Fe. Taos County: 4.8 km E Questa; Frazier Mt., Twining; Williams Lake Trail, Twining; Gold Hill near Red River; just E of Rio Puebla; Red River Pass; trail from Red River to Wheeler Peak. Valencia County: Grants, Mt. Taylor Summit (3353 m); Mount Taylor. North Dakota: Benson County: no specific location. Billings Coun- ty: T.R. National Memorial Park. Dunn County: T146-R97-S25-P400; Killdeer Mountains. Mc- Kenzie County: N unit T.R. Park; T146-R98-S16-P110. Mclean County: Washburn Park; 6.4 km S Washburn Rest Area. Mercer Coun- ty: Hazen. Monton County: no specific location. Pembina County: no specific location. Slope Coun- ty: Chalky Buttes; T136-R102-S14-P200; Burning Coal Vein. Williams County: no specific location. Oregon: Benton County: Corvallis Entomological Research Farm; Marys Peak Parker Creek, near Marys Peak Campground. Grant County: Malheur National Forest, Blue Mountain Hot Springs. Har- ney County: 24 km S Bums; Steen Mountains. Jef- ferson County: head of Metolius River, Riverside Forest Camp. Union County: Insler, Harris Moun- tains. Yamhill County: McMinnville. South Dakota: HOLMBERG & COKENDOLPHER— RE-DESCRIPTION OF TOGWOTEEUS BICEPS 233 Custer County: Custer State Park. Lawrence Coun- ty: Custer Park; 5.5 km S Deadwood on Highway 85 A; Spring Creek Camp, 17.7 km NE Hill City. Pennington County: Mount Rushmore. Utah: Coun- ty?: Blacksmith Fork Canyon; Butterfield Canyon, Oquirrh Mountains; Othess Mantes Canyon; Puffer Lake. Beaver County: Beaver Canyon, 16 km (di- rection?) from Beaver City; Kents Lake Camp, 25.8 km E Beaver City. Box Elder County: Bear River City; Clear Creek, Raft River Mountains; Dove Creek, Raft River Mountains; Rabbit Springs, 9.7 km N Lucin; Raft River Mountains, 12.9 km S Lynn. Cache County: Beaver Mountain, Wasatch Range; Logan Canyon; Logan River; Red Banks, Logan Canyon (1829 m). Dagget County: Hideout Canyon; Junction Deep and Carter Creek. Emery County: Ferron Reservoir; Huntington Canyon. Garfield or Wayne County: Horse Valley, Henry Mountains. Garfield County: Aquarius Plateau, Steep Creek; Blue Spruce Camp, 29 km N Escalan- te (2438 m); Wild Cat Ranger Station, 24.2 km N Boulder. Grand County: La Sal Mountains; Mirror Lake, Uintah Mountains; Warner Ranger Station, 45.1 km ESE Moar (2804 m). Iron County: Cedar Breaks. Juab County: Trout Creek. Morgan County: Bells Canyon. Millard County: Oak Creek Camp, 14 km E Oak City. Rich County: Bear Lake (east side). Salt Lake County: 6.4 km up City Creek Can- yon, Salt Lake City; City Creek, Salt Lake City; Emigration Canyon; Mill Creek Canyon; Mill Creek, Salt Lake City; Salt Lake City. San Juan County; Buckboard Flat Camp, 11.3 km W Mon- ticello (2682 m); Dalton Springs Camp, 8 km W Monticello (2591 m); La Sal Mountains. Sanpete County: Moroni. Sevier County: Fish Lake. Summit County: Beaver Canyon; Hoop Lake, Uintah Moun- tains. Tooele County: Loop Camp, 20.9 km SW Grants ville (2256 m); South Willow Canyon. Uin- tah County: Iron Springs Camp, 40.3 km N Vernal (2652 m); Kaler Hollow Camp, 35.4 km NNW Ver- nal (2713 m). Utah County: American Fork Can- yon, Timponogos; Aspen Grove; Timpanogos. Wa- satch County: Provo River at North Fork. Washington County: Pine Valley Mountains; Zion National Park. Washington: Kittitas County: El- lensberg. Klickitat County: near Maryhill, 11.3 km E Daller Ferry; Trout Lake County Park. Stevens County: 4.8 km N, 12.8 km NE and 40 km N Wellpinit. Whitman County: Elberton. Wyoming: County?: Freemouth. Albany County: Medicine Bow Peak, near Centennial; Woods Landing. Big Horn County: Porcupine Camp, 53.7 km E Lovell. Carbon County: Bottle Creek Camp, 11.3 km SW of Encampment; 12.9 km SW of Encampment (2621 m). Converse County: Medicine Bow Moun- tains; Summit Laramire Mountains near Pole Mountains. Crook County: Reuter Canyon Camp, 8 km NW Sundance (1737 m); 8 km N Sundance. Johnson County: Bighorn National Forest, Circle Park Rec. Area. Laramie County: Cheyenne. Lin- coln County: Cokeville. Park County: Lake Creek Camp, 20.9 km SE Cooke; Lost Creek Camp; Mount Washburn Summit, Yellowstone National Park. Sheridan County: Ranger Creek Camp, 30.6 km SW Big Horn (2377 m). Sublette County: Low- er Green River Lake, Wind River Range (2438 m). Sweetwater County: Green River. Teton County: Grand Teton National Park; near Moran; Moran Junction; Owl Creek Headquarters, 48 km N Jack- son; Stewart River Station; spring runs crossing Route 287 in Togwotee Pass; Teton Pass; Wilson; Old Faithful, Yellowstone National Park; near Yel- lowstone Lake, Yellowstone National Park. Description.-— Body with thick, hard, tu- berculate-microgranulate cuticle on dorsal surface (Figs. 2, 3); off-center micropores present on dorsal tubercles (Figs. 5, 6); base coloration varies from amber to black, with dorsal specks or patterns of lighter color sometimes present; dorsum sometimes with faint central figure; males tend to be more sclerotized and have more denticles and there- fore appear darker than females; preocular area without mound but with two groups of small denticles near anterior margin edge (Fig. 4, arrow b); supracheliceral lamellae well de- veloped and toothed (Fig. 4, arrow a); ocular tubercle low, rounded, covered by many prominent spines (Fig. 3); darker ring often encircling each eye; light area extends be- tween the eyes and from the ocular tubercle to the anterior edge of the prosoma and usu- ally forms a distinct bifid stripe (Fig. 2). Ab- domen with faint indication of segmentation dorsally. Genital operculum without notice- able crest ventrally; with many microscopic spicules dorsally (Figs. 7-9). Chelicerae not enlarged, without apophyses on jaws, ventral spur on basal segment large and covered with many spicules; with 4-6 slit sensilla on sec- ond segment (Figs. 18, 21). Pedipalps without apophyses on distal ends of patellae or tibiae in juveniles or adults, distal end of femora without campaniform organ (slit-sensillum present), claw smooth, not toothed (Figs. 13, 14, 17); pedipalps sexually dimorphic; male pedipalps modified, enlarged, tarsus bulbous at base and with two rows of ventral denticles; midventral area of tibia slightly compressed (most noticeable on mesal side) (Figs. 14-16, 36, 37). Legs generally short and wide; femur I as wide or wider than ocular tubercle, fem- ora I usually equal to or shorter than body length, no pseudoarticulary nodules in femora. 234 THE JOURNAL OF ARACHNOLOGY Figures 2-6. — Body and integument of Togwoteeus biceps. 2, Dorsal view of adult male (non-SEM); 3, Ocular tubercle and anterior portion of cephalothorax of male; 4, Supracheliceral lamella (arrow a) and anterior portion of cephalothorax of female with tubercles (arrow b); 5, Cuticle of prosoma of male; 6, Detail of cuticle. Scales = 0.5 mm in Figs. 2, 3; 0.1 mm in Fig. 4; 0.05 mm in Fig. 5; 10 jjim in Fig. 6. tibiae II with (longer legged specimens) or without (lectotype and shorter legged speci- mens) pseudosegments; tibiae I, III, IV with- out pseudosegments; femora, patellae, tibiae with randomly spaced (without rows) pointed tubercles (Fig. 10); patellae and distal tips of leg femora and tibiae often darkly shaded; each leg coxa with center spine dorsally, with at most a weakly developed lateral row of denticles (Fig. 7). Penis alate (Figs. 22-25, 31-33); ca. 5 p.m longitudinal slit near the tip of the stylus of the penis glans (Fig. 24). Ovi- HOLMBERG & COKENDOLPHER— RE-DESCRIPTION OF TOGWOTEEUS BICEPS 235 Figures 7-12. — Morphology and anatomy of Togwoteeus biceps. 7, Anteroventral view of anterior portion of male; 8, Spicules on inner lining of the anterior edge of male genital operculum; 9, Detail of operculum spicules; 10, Lateral view of leg I of female; 11, Detail of tarsal leg pores of female; 12, Tip of leg tarsus with smooth claw of male. Scales = 1 mm in Fig. 7; 0.1 mm in Figs. 8, 10, 12; 0.5 mm in Fig. 11; 10 |jLm in Fig. 9. positor relatively long, 34 segmented in par- alectotype; with four slit-sensilla (two dorsal, two ventral) per lateral half; 3 segmented fur- ca (Figs. 26, 27, 35). Ovipositor enclosed in two sheaths, details as in Figs. 28—30. Seminal receptacles as in Fig. 34, located in segments 3-4 of ovipositor (Fig. 35). Body measurements. — The results of mea- suring 80 males and 74 females are given in Table 2. No simple measurement dine was observed between the four areas (Table 1) and the data were pooled. Female body measure- ments are larger than those of males. Male pedipalps measurements are larger than fe- 236 THE JOURNAL OF ARACHNOLOGY Figures 13-17. — Right pedipalp of Togwoteeus biceps. 13, Lateral view of female; 14, Detail of male tarsus; 15, Denticles and pores on tarsus of male; 16, Detail (3.4X enlargement) from insert in Fig. 15; 17, Smooth claw of male. Scales = 0.5 mm in Fig. 13; 0.1 nun in Fig. 14; 0.05 mm in Figure 17; 50 |jLm in Fig. 15. males, except for palpal tarsus length (which is longer in females) and width (which is the same for both sexes). Male leg measurements were generally longer except for femur II and IV. The length of the genital operculum is lon- ger in males but wider at the base in females. The neck width is the same. Male and female measurements for the ocular tubercle are the same except that the female’s is slightly closer to the anterior margin. The mode numbers of metatarsal bands for both sexes are four; the range varies between 2-10. Ultrastructure. — The pedipalp sense or- gans of Togwoteeus appear similar to those observed by Spicer (1987) in other harvest- men, i.e., Leiobunum C. Koch 1839 and Eu- mesosoma Cokendolpher 1980. Like Spicer, we found sensilla trichodea and chaetica on the pedipalps of T. biceps (Figs. 11, 12, 14, 17). The “tarsal organs” of Spicer were also observed on the ventral surface of the pedi- palp tarsus (Figs. 15, 16). The prosomal dor- sum revealed a tuberculate-microgranulate morphology (Figs. 3-4). The prominent tu- bercles have asymmetrical arms, with the cen- tral regions containing off-center micropores. The abdominal setae arise from tubercles el- evated from the surface of the integument. These tubercles appear to be constricted at their bases. The dorsal morphology of T. bi- ceps is unlike any other harvestman thus far examined {cf. Murphree 1988). The off-center HOLMBERG & COKENDOLPHER— RE-DESCRIPTION OF TOGWOTEEUS BICEPS 237 Figures 18-21. — CheUcerae of female Togwoteeus biceps. 18, Lateral view; 19, Lateral view of basal tooth with numerous spicules (close-up of Fig. 18); 20, Mesal view; 21, Detail of slit sense organ at base of movable cheliceral jaw (close-up of Fig. 18). Scales = 0.5 mm in Figs. 18, 20; 0.05 mm in Figs. 19, 21. 238 THE JOURNAL OF ARACHNOLOGY Figures 22-21 . — Genital structures of Togwoteeus biceps. 22, Ventrolateral view of distal end of penis; 23, Lateral view of glans of penis; 24, Lateral view of stylus of penis - arrow indicates location of a 5 (xm slit; 25, Dorsal view of stylus of penis; 26, Distal end of ovipositor with 3-segmented furca; 27, Distolateral view of sensillae on ovipositor. Scales = 0.1 mm in Figs. 22, 26, 27; 0.05 mm in Figs. 23-25. placement of micropores is unlike the central placement of members of Leiobunum and Hadrobunus Banks 1900. Eumesosoma appar= ently does not have micropores atop of dorsal tubercles (Murphree 1988: Fig. 18). The ul- trastructure of the ovipositor sheaths are here- in illustrated for the first time as are some de- tails of the penis and ovipositor. Variation. — We occasionally found speci- mens that had leg lengths nearly double the HOLMBERG & COKENDOLPHER— RE-DESCRIPTION OF TOGWOTEEUS BICEPS 239 Figures 28-30. — Ovipositor of Togwoteeus biceps. 28, Mid-section of ovipositor and broken sheaths, o = outer and i = inner sheaths; 29, Expanded view of inner sheath of ovipositor; 30, Expanded view of cross-section of inner sheath. Scales = 0.1 mm in Fig. 28, 10 jxm in Figs. 29, 30. normal. For example, a female specimen from near Osyoos, British Columbia had a femur II length of 9.60 mm {cf. mean of 5.35 mm, Ta- ble 2). However after examining other char- acteristics and measuring many individuals, we concluded that these were exceptional in- dividuals and not another species. Goodnight & Goodnight (1953) stated that a north-south dine in coloration could be demonstrated for this species. The lighter colored individuals were in the north and the darker colored in- dividuals were in the south. While this may be partially true, it is not as simple as that. It appears age and sex of the animal may play a significant role in the color of the animal. Ex- amination of series collected through time re- vealed older animals and males are darker. The role of elevation and moisture have not been investigated, but as evidenced in other arthropods they may play a role in the deter- mination of pigmentation. There was no ob- vious dine in the morphological measure- ments of the specimens examined in this study. Anomalies. — Holmberg & Kokko (1983) reported on an eye-less T. biceps. It had no ocular tubercle and only degenerate optic nerve masses. This anomaly was found in only one specimen. During this study some abnormalities were noticed in the formation of the denticles near the anterior edge of the cephalothorax and the supracheliceral lamel- lae. Usually one such structure was smaller than normal and with fewer and smaller den- ticles. Chromosomes. — Karyotypes from two subadult males and one subadult female re- vealed 2n = 22, all being metacentric chro- mosomes (Fig 38). Sex chromosomes were not detected in the male karyotypes. The chro- mosomes in the two karyotypes from the fe- male were not condensed and sex chromo- somes could not be distinguished. Studies detailing chromosomes of harvestmen are few. The known counts/karyotypes were pre- sented or reviewed by Tsurusaki & Coken- dolpher (1990) and Cokendolpher & Jones (1991). Including the present study, karyo- types of 40 species of the superfamily Phal- angioidea are known. The diploid chromo- some numbers thus far known for the superfamily range between 10 and 36, with 22 being recorded in several unrelated genera of the Protolophidae (Protolophus Banks 1893) 240 THE JOURNAL OF ARACHNOLOGY Figures 31-37. — Anatomy of Togwoteeus biceps (male lectotype, female paralectotype). 31, Ventral view of penis; 32, Lateral view of penis; 33, Dorsal view of distal end of penis; 34, Seminal receptacles; 35, Distal end of ovipositor with seminal receptacles (dotted lines); 36, Mesal view of male pedipalp; 37, Lateral view of male pedipalp. fl = femur length, tl = tibia length. and Sclerosomatidae {Dalquestia Cokendol- pher 1984, Eumesosoma, Gagrellula Roewer 1911, Leiobunum). Habitat. — This species is found in many habitats. In the mountains specimens are often found in densely wooded areas as well as on windswept mountain-tops above the tree line. They also occur in many dry habitats, but of- ten near water bodies, especially in northern prairies. In the southern part of their range, they are restricted to higher elevations. They have been found under rocks, logs, and other ground debris. A few individuals were also obtained in deserted buildings. Only rarely have they been obtained by sweeping vege- tation. They do not appear to aggregate in pro- tected sites like many other sclerosomatids. Phenology.— Although the collection data HOLMBERG & COKENDOLPHER— RE-DESCRIPTION OF TOGWOTEEUS BICEPS 241 Table 2„ — Morphological measurements (mm) and counts of Togwoteeus biceps. Data pooled from specimens described in Table L SE = Standard error, ns = not significant, probability > 0.05. Statistically significant larger values in bold. Males Females Body part Mean (SE) Range n Mean (SE) Range n P Body Prosoma width 2.68 (0.031) 2.08-3.12 80 2.87 (0.035) 2.04-3.40 74 <0.001 Body length 4.84 (0.053) 3.60-5.76 80 6.16 (0.084) 3.72-7.48 74 <0.001 Abdomen width 2.89 (0.036) 1.88-3.08 80 3.58 (0.049) 2.28-4.92 74 <0.001 Abdomen height 2.51 (0.029) 1.88-3.08 80 3.40 (0.061) 1.88-4.68 74 <0.001 Pedipalps Femur length 1.15 (0.015) 0.90-1.36 80 0.99 (0.013) 0.72-1.20 74 <0.001 Femur width 0.33 (0.005) 0.24-0.44 80 0.25 (0.003) 0.16-0.30 74 <0.001 Patella length 0.58 (0.006) 0.44-0.66 80 0.48 (0.006) 0.36-0.58 74 <0.001 Patella width 0.35 (0.004) 0.26-0.40 80 0.30 (0.003) 0.22-0.36 74 <0.001 Tibia length 0.81 (0.010) 0.62-1.00 80 0.70 (0.009) 0.48-0.88 74 <0.001 Tibia width 0.34 (0.004) 0.24-0.44 80 0.26 (0.003) 0.20-0.34 74 <0.001 Tarsus length 1.06 (0.013) 0.82-1.28 80 1.12 (0.014) 0.82-1.38 74 0.002 Tarsus width 0.17 (0.004) 0.12-0.24 80 0.17 (0.003) 0.10-0.22 74 0.23 ns Legs Femur I length 3.22 (0.087) 2.08-4.92 79 2.90 (0.072) 1.76-4.56 74 <0.006 Tibia I length 2.59 (0.066) 1.60-3.84 78 2.28 (0.051) 1.52-3.40 73 <0.001 Femur II length 5.76 (0.174) 3.64-9.44 80 5.35 (0.156) 3.36-8.52 74 >0.08 ns Tibia II length 5.08 (0.156) 3.08-8.16 80 4.16 (0.130) 2.88-7.32 74 0.02 Femur III length 3.45 (0.089) 2.24-5.20 80 3.13 (0.072) 2.04-4.56 74 0.006 Tibia III length 2.71 (0.068) 1.72-3.80 80 2.42 (0.053) 1.60-3.56 74 0.001 Femur IV length 5.11 (0.137) 3.40-7.76 79 4.86 (0.116) 3.20-7.60 74 >0.16 ns Tibia IV length 3.80 (0.095) 2.52-5.60 77 3.49 (0.080) 2.32-5.40 74 0.015 Genital operculum Length 3.13 (0.037) 2.48-3.80 80 2.92 (0.036) 2.16-3.48 74 <0.01 Neck width 1.28 (0.013) 0.99-1.52 80 1.29 (0.015) 0.96-1.52 74 0.56 ns Base width 2.25 (0.026) 1.72-2.88 80 2.39 (0.026) 1.76-2.92 74 <0.001 Ocular tubercle To anterior margin 0.52 (0.007) 0.38-0.64 80 0.49 (0.007) 0.36-0.64 74 <0.004 Length 0.45 (0.004) 0.38-0.54 80 0.45 (0.006) 0.36-0.56 74 0.66 ns Width 0.43 (0.003) 0.38-0.50 80 0.43 (0.004) 0.34-0.50 74 0.36 ns Height 0.26 (0.005) 0.16-0.32 80 0.26 (0.004) 0.20-0.32 74 0.62 ns Metatarsal II bands 4.82 (0.146) 2-8 4.53 (0.183) 2-10 73 0.35 ns (Table 3) are biased (i.e., most collection dates in the summer, most specimens collected were larger — usually adults, most collection sites between 40--50°N latitude), it appears that T. biceps overwinters as immatures which be- come adults in May or June and then die by fall. It is likely that the majority of the indi- O O «S viduals have a one year life cycle, but it is possible that late maturing adults may produce offspring that take two summers to reach ma- turity. There is no evidence that this phenol- ogy pattern changes over the latitudinal range of the species. Parasites. — Poinar (1985) reported an un- « t» an f t t# Figure 38. — Karyotype (2n = 22) of subadult male Togwoteeus biceps from near Logan, Utah. 242 THE JOURNAL OF ARACHNOLOGY Table 3. — Latitude versus time of year for collections of Togwoteeus biceps. I = immatures, A = adults. Note that 98% of the July records for the latitude 30-34° grouping came from pit traps from one locality. The label date is 3 July but probably most of the collections were from June. Spring Summer Fall Winter Mar. Apr. May June July Aug. Sep. Oct. Nov. Dec. Jan. Feb. Latitude I, A I, A I, A I, A I, A I, A I, A I, A I, A I, A I, A I, A 50-55° 4, 0 21,21 10, 17 7, 28 0, 1 L 2 L 1 45-49° 35, 0 47, 0 39, 69 93, 497 11, 493 2, 73 2, 1 14, 4 23, 2 1, 0 5, 0 19, 0 40-44° 3, 0 38, 4 31, 40 25, 63 15, 77 11, 30 0, 1 35-39° 12, 0 25, 23 6, 89 1, 22 6, 20 0, 2 30-34° 277, 181 0, 1 % adults 0 0 46 78 72 91 73 35 9 0 0 0 identified juvenile mermithid nematod (Aga- momermis sp.) parasite from this harvestman. Cokendolpher (1993) recorded unidentified Leptus sp. mites from T. biceps. CONCLUSIONS It appears that Togwoteeus biceps is mono- typic. The range of the species extends through much of the western prairie and mountain areas of Canada and USA. The lat- itudinal range is about 33-54°N, longitudinal about 98-1 24°W. This species has the greatest elevational range (<500 to 4100 m) and oc- curs at the highest elevation of any recorded harvestman in North America. ACKNOWLEDGMENTS We thank Mr. Gerald Hilchie and Dr. Jim Ryan, for making the many measurements and for other assistance; Mr. Richard Powell for help with the statistical programs, and Ms. Edith Schwaldt and Mr. George Braybrook for help with electron microscopy. Much of this project’s costs were funded by Athabasca Uni- versity. Texas Tech University, Lubbock, par- tially supported JCC during the early years of this project (1980-1986). We thank the following individuals for checking their collections for the types of '^Homolophus” punctatus and "'Homolophus” biceps: Mr. Rodney L. Crawford (Thomas Burke Memorial Washington State Museum, University of Washington, Seattle); Dr. Jurgen Gruber (Naturhistorisches Museum Wien, Wien); the late Dr. Clarence J. Goodnight (Kalamazoo); Dr. Daniel Otte (Academy of Natural Sciences, Philadelphia); Dr. Gisela Rack (Zoologisches Insitut und Zoologisches Museum, Hamburg); the late Dr. Vladimir Sil- havy (Treble); and Dr. Fred Wanless (British Museum of Natural History, London). These types were also searched for by numerous oth- er curators listed below. The following curators and individuals loaned us material for study: Dr. Gianna Ar- bocco (Museo Civico di Storia Naturale “Gia- como Doria”, Genova); Dr. Edward U. Bals- baugh, Jr. (North Dakota State University, Fargo); Dr. David Barr (Royal Ontario Museum, Toronto); Dr. Philip D. Bragg (pers. coll., Vancouver, British Columbia); Mr. Don J. Buckle (pers. coll.. Saskatoon, Saskatche- wan); Dr. Robert A. Cannings (Royal British Columbia Museum, Victoria); Dr. R.E. Cra- bill, Jr. and Dr. Jonathan Coddington (National Museum of Natural History, Smithsonian In- stitution, Washington, D.C. ); Dr. Charles D. Dondale and Mr. James H. Redner (Canadian National Collections of Insects, Arachnids and Nematodes, Ottawa, Ontario); Dr. Albert T. Finnamore and Mr. Terry Thormin (Provincial Museum of Alberta, Edmonton); Mr. Saul Frommer (UCR Entomological Teaching and Research Collection, University of California, Riverside); Dr. Manfred Grasshoff (Sencken- berg Natur-Museum, Frankfurt am Main); Dr. David H. Kavanaugh and Mr. Vincent E Lee (California Academy of Sciences, San Fran- cisco); Dr. John B. Kethley (Field Museum of Natural History, Chicago); Dr. Torbjom Kro- nestedt (Naturhistoriska Riksmuseet, Stock- holm); Mr. Frank W. Merikel (University of Idaho, Moscow); Dr. Auturo Munoz-Cuevas (Museum National d’Historie Naturelle, Par- is); Dr. Norman 1. Platnick (American Muse- um of Natural History, New York); Mr. James R. Reddell (Texas Memorial Museum, The HOLMBERG & COKENDOLPHER— RE-DESCRIPTION OF TOGWOTEEUS BICEPS 243 University of Texas, Austin); Dr. G.G.E. Scudder and Dr. Syd G. Cannings (University of British Columbia, Vancouver); Dr. William A. Shear (pers. coll., Hampden-Sydney, Vir- ginia). We also thank the many amateur and professional arachnophiles who made the col- lections that ended up in the above mentioned institutions. LITERATURE CITED Banks, N. 1893. The Phalanginae of the United States. Canadian Entomol., 25:205-211. Banks, N. 1894. Washington Phalangida, with de- scription of a new southern Liobunum. Canadian Entomol., 26:160-164. Banks, N. 1895. The Arachnida of Colorado. Ann. New York Acad. Sci., 8:417-434. Banks, N. 1900. Arachnida. P. 123, In Some in- sects of the Hudsonian Zone in New Mexico. (T.D.A. Cockerell, ed.). Psyche, 9:123-126. Banks, N. 1901. Synopses of North- American In- vertebrates. XVI. The Phalangida. American Nat., 35:669-679. Banks, N. 1902. Some Arachnida from New Mex- ico. Proc. Acad. Nat. Sci., Philadelphia, (1901) 53:568-597. Banks, N. 1916. Report on Arachnida collected by Messrs. Currie, Caudell, and Dyar in British Co- lumbia. Proc. United States Nat. Mus., 51:67-72. Blake, I.H. 1945 An ecological reconnoissance [sic\ in the Medicine Bow Mountains. Ecol. Monogr., 15(3):209-242. Bragg, P.D. & R.E. Leech. 1972. Additional re- cords of spiders (Araneida) and harvestmen (Phalangida) from British Columbia. J. Entomol. Soc. British Columbia, 69:67-71. Cockerell, T.D.A. 1907. Some Coleoptera and Arachnida from Florissant, Colorado. Bull. American Mus. Nat. Hist., 23:617-621. Cockerell, T.D.A. 1911. The fauna of Boulder County, Colorado. Univ. Colorado Stud., 8:227- 256. Cokendolpher, J.C. 1981. Revision of the genus Trachyrhinus Weed (Opiliones, Phalangioidea). J. ArachnoL, 9:1-18. Cokendolpher, J.C. 1985. Revision of the harvest- man genus Leptobunus and dismantlement of the Leptobunidae (Arachnida: Opiliones: Palpa- tores). J. New York Entomol. Soc., 92:371-402. Cokendolpher, J.C. 1987. On the identity of the genus Homolophus: a senior synonym of Eu- phalangium (Opiliones: Phalangiidae). Acta Ar- achnoL, 35:89-96. Cokendolpher, J.C. 1993. Pathogens and parasites of Opiliones (Arthropoda: Arachnida). J. Arach- noL, 21:120-146. Cokendolpher, J.C. & J.D. Brown. 1985. Air-dry method for studying chromosomes of insects and arachnids. Entomol. News, 96:114-118. Cokendolpher, J.C. & J.E. Cokendolpher. 1982. Reexamination of the Tertiary harvestmen from the Florissant Formation, Colorado (Arachnida: Opiliones: Palpatores). J. Paleontol., 56:1213- 1217. Cokendolpher, J.C. & S.R. Jones. 1991. Karyotype and notes on the male reproductive system and natural history of the harvestman Vonones sayi (Simon) (Opiliones, Cosmetidae). Proc. Entomol. Soc. Washington, 93(1): 86-91. Cokendolpher, J.C. & V.E Lee. 1993. Catalogue of the Cyphopalpatores and bibliography of the har- vestmen (Arachnida, Opiliones) of Greenland, Canada, U.S.A., and Mexico. Privately pub- lished, Lubbock. Comstock, J.H. 1912. The Spider Book. Double- day, Page & Co., New York. Comstock, J.H. 1940. The Spider Book. (WJ. Gertsch, ed.) Doubleday, Doran & Co., New York. Crawford, R.L. 1992. Catalogue of the genera and type species of the harvestman superfamily Phal- angioidea (Arachnida). Burke Mus. Contrib. An- thropol. Nat. Hist., No. 8, 60 pp. Edgar, A.L. 1990. Opiliones (Phalangida). Pp. 529-581. In Soil Biology Guide. (D.L. Dindal, ed.). John Wiley & Sons, New York. Goodnight, C.J. & M.L. Goodnight. 1953. Taxo- nomic recognition of variation in Opiliones. Syst. Zool., 2:173-180. Holmberg, R.G. 1970. The odoriferous glands of some Palpatores Phalangida (Opiliones) (Arach- nida). M.Sc. Thesis, Univ. of Saskatchewan, Sas- katoon. Holmberg, R.G., P.D. Bragg & J. Belicek. 1981. Opiliones (Arachnida) of western Canada: a progress report. Proc. 28th Annu. Meet. Ento- mol. Soc. Alberta, (1980) 28:19. Holmberg, R.G. & E.G. Kokko. 1983. A blind Homolophus biceps (Arachnida: Opiliones). En- tomol. News, 94(2):49-52. Kastner, A. 1937. Chelicerata. 7. Ordnung der Arachnida: Opiliones Sundeval = Weberknechte. Pp. 300-393 In Handbuch der Zoologie. Vol. 3, No. 2. (W. Kukenthal & T. Krumbach, eds.). Walter de Gruyter & Co., Berlin & Leipzig. Katayama, R.W. & R.L, Post. 1974. Phalangida of North Dakota. North Dakota Insects Publ., No. 9. Levi, H.W & L.R. Levi. 1951. Report on a collec- tion of spiders and harvestmen from Wyoming and neighboring states, Zoologica, 36:219-237. Levi, H.W & L.R. Levi. 1968. Invertebrate Zool- ogy, Arachnids and Myriapods. Vol, II. Trans, and adapted from Kaestner (1937). John Wiley & Sons, Inc., New York. Levi, L.R. & H.W. Levi. 1955. Spiders and har- 244 THE JOURNAL OF ARACHNOLOGY vestmen from Waterton and Glacier National Parks. Canadian Field-Nat., 69(2):32-40. Murphree, C.S. 1988. Morphology of the dorsal integument of ten opilionid species (Arachnida, Opiliones). J. Arachnol., 16:237-252. Pavesi, P. 1889. Un nuovo Nemastomidae Ameri- cano. R. 1st. Lombardo Sci. Lett. Rend., Milan, (2)32:530-533. Poinar, Jr., G.O. 1985. Mermithid (Nematoda) par- asites of spiders and harvestmen. J. Arachnol., 13:121-128. Roewer, C.E 1910. Revision der Opiliones Pla- giostethi (= Opiliones Palpatores) 1. Teil: Fami- lie der Phalangiidae. (Subfamilien: Gagrellini, Liobunini, Leptobunini.). Abh. Geb. Naturwiss., Naturwiss. Ver, Hamburg, 19(4): 1-294. Roewer, C.F. 1912. Revision der Opiliones Palpa- tores (= Opiliones Plagiostethi) 11. Teil: Familie der Phalangiidae. (Subfamilie: Sclerosomini, Oli- golophini, Phalangiini). Abh. Geb. Naturwiss., Naturwiss. Ver. Hamburg, 20(1): 1-295. Roewer, C.F. 1923. Die Weberknechte der Erde. Systematische Bearbeitung der bisher bekannten Opiliones. Gustav Fischer, Jena. Roewer, C.F. 1929. Die Opilioniden der schwed- ischen Kamtchatka-Expedition 1920-1929. Ark. ZooL, 21B(2):l-3. Roewer, C.F. 1952. Einige Phalangiiden aus den Vereinigten Staaten von Nord-Amerika. Zool. Anz., 149:267-273. Roewer, C.F. 1957. Uber Oligolophinae, Caddoin- ae, Sclerosomatinae, Leiobuninae, Neopilioninae und Leptobuninae (Phalangiidae, Opiliones Pal- pators). (Weitere Weberknechte XX). Sencken- bergiana Biol., 38:323-354. Roewer, C.F. 1960. Solifugen und Opilioniden, Araneae, Orthognathae, Haplogynae und Ente- legynae (Contribution a 1’ etude de la fauna d’ Afghanistan 23). Goteborgs Vetensk. Samhal- les Handl. (Ser. 6B), 8(7): 1-53. Schmoller, R.R. 1970. Life histories of alpine tun- dra Arachnida in Colorado. American Midi. Nat., 83:119-133. Schmoller, R.R. 1971a. Habitats and zoogeography of alpine tundra Arachnida and Carabidae (Co- leoptera) in Colorado. Southwest. Nat., 15:319- 329. Schmoller, R.R. 1971b. Nocturnal arthropods in the alpine tundra of Colorado. Arctic Alp. Res., 3:345-352. Spicer, G.S. 1987. Scanning electron microscopy of the palp sense organs of the harvestman Leiobunum townsendi (Arachnida: Opiliones). Trans. American Microsc. Soc., 106:232-239. Thorell, T. 1877. Descriptions of the Araneae col- lected in Colorado in 1875 by A.S. Packard, Jr., M.D. Bull. United States Geol. Geogr. Surv. Terr., 3:477-529. Tsurusaki, N. & J.C. Cokendolpher. 1990. Chro- mosomes of sixteen species of harvestmen (Arachnida, Opiliones, Caddidae and Phalangi- idae). J. Arachnol., 18:151-166. Manuscript received 5 September 1996, accepted 13 March 1997. 1997. The Journal of Arachnology 25:245-250 NOTES ON THE TAXONOMY OF SOME OLD WORLD SCORPIONS (SCORPIONES: BUTHIDAE, CHACTIDAE, ISCHNURIDAE, SCORPIONIDAE) Victor Fet: Department of Biological Sciences, Marshall University, Huntington, West Virginia 25755-2510 USA ABSTRACT. The following new synonymies are found among the scorpions of the Old World: Buthoscorpio Werner 1936 = Pocockius Francke 1985, NEW SYNONYMY; Androctonus (Leiurus) quinquestriatus Ehren- berg 1828 = Androctonus (Liurus) quinquestriatus aculeatus Ehrenberg 1831, NEW SYNONYMY; Euscorpius mingrelicus gamma (Caporiacco 1950) (in part), NEW COMBINATION = Euscorpius mingrelicus caprai Bonacina 1980, NEW SYNONYMY The following new homonymies are pubUshed: Buthus (Hottentotta) hebraeus Werner 1935, NEW HOMONYMY; Buthus acutecarinatus judaicus Birula 1905, NEW HOMON- YMY; Euscorpius germanus polytrichus Hadzi 1929, NEW HOMONYMY; Euscorpius germanus oligotrichus Hadzi 1929, NEW HOMONYMY; Euscorpius carpathicus oligotrichus Hadzi 1929, NEW HOMONYMY; Scorpio maurus subtypicus Birula 1910, NEW HOMONYMY. The following new replacement names are introduced to replace junior homonyms: Androctonus amoreuxii levyi Fet, NOMEN NOVUM = Buthus {Hot- tentotta) hebraeus Werner 1935, NEW SYNONYMY; Lychas marmoreus lucienkochi Fet, NOMEN NOVUM = Lychas marmoreus obscurus Kraepehn 1916, NEW SYNONYMY; Uroplectes fischeri caporiaccoi Fet, NOMEN NOVUM = Uroplectes fischeri intermedius Caporiacco 1941, NEW SYNONYMY; Hadogenes tri- chiurus wemeri Fet, NOMEN NOVUM = Hadogenes trichiurus paucidens Werner 1939, NEW SYNONYMY; Scorpio maurus birulai Fet, NOMEN NOVUM = Scorpio maurus subtypicus Birula 1910, NEW SYNON- YMY. Type species are designated for Pilumnus C.L. Koch 1837, Repucha as proposed by Francke (1985), and two subgenera of Pandinus ThoreU as proposed by Vachon (1974). An incorrect original spelling Tricho- buthus grubleri Vachon 1941 is corrected to T. guebleri (a junior synonym of Buthiscus bicalcaratus Birula 1905). The correct date of the description for Pectinibuthus Fet is 1984, but not 1987. The name Euscorpius carpathicus mesotrichus Hadzi 1929 is a junior homonym and cannot be used. These notes discuss some taxonomic prob- lems within the Old World scorpiofauna. This fauna, especially in the Mediterranean area and Middle East, has been described intensively since the 18th century, often creating multiple synonymies and homonymies. A considerable effort of many generations of scorpion taxono- mists (including such prominent names as T. ThoreU, R.L Pocock, E. Simon, K. Kraepehn, A. Birula, M. Vachon, O. Francke, W.R. Lour- engo) led to the clarification of many nomen- clatural problems; however, a number of situa- tions exist which do not comply with the International Code of Zoological Nomenclature (ICZN 1985; further quoted as separate Arti- cles). Below, I attempted to analyze those sit- uations. FAMILY BUTHIDAE C.L. KOCH 1837 Genus Androctonus Ehrenberg 1828 Buthus {Hottentotta) hebraeus Werner 1935, NEW HOMONYMY (currently Androctonus amoreuxii hebraeus) is found to be a primary junior homonym of Buthus quinquestriatus he- braeus Birula 1908 (currently Leiurus quin- questriatus hebraeus) (Article 53 of the Code), and thus is permanently invahd (Article 52) and has to be replaced. A new replacement name is introduced, Androctonus amoreuxii levyi Fet, NOMEN NOVUM = Buthus {Hottentotta) he- braeus Werner 1935, NEW SYNONYMY. Et- ymology: a patronym honoring the Israeh ar- achnologist Dr. Gershom Levy. Genus Buthiscus Birula 1905 Vachon (1941) described Trichobuthus grubleri from Algeria. Later (Vachon 1942) he found this species to be a junior synonym of Buthiscus bicalcaratus Birula 1905. The original patronym grubleri was supposed to be formed from the collector’s name, M. Gtib- ler, and is therefore an incorrect original spell- ing [Article 32(c)]. Vachon (1952) admitted 245 246 THE JOURNAL OF ARACHNOLOGY clear evidence of an inadvertent error {lapsus calami). It is here corrected to Trichobuthus guebleri Vachon 1941; following Article 32(d)(i)(2), the letter “e” is inserted after the vowel. Genus Buthoscorpio Werner 1936 This genus was originally described in the family Scorpionidae, with a single (type by monotypy) species Buthoscorpio laevicauda Werner 1936, from India. Vachon (1961) found this species to be a buthid, and a junior synonym of Stenochirus politus Pocock 1899 (Buthidae). Thus, the name Buthoscorpio Werner 1936 became a junior synonym of Stenochirus Karsch 1892 (type species by original designation Stenochirus sarasinorum Karsch 1892, from Sri Lanka). Later, Francke (1985) discovered that Karsch’s name was a junior homonym of Stenochirus Oppel 1862 (Crustacea). A re- placement name, Pocockius Francke 1985, was introduced as a nomen novum for Steno- chirus Karsch 1892. However, the name Buth- oscorpio Werner 1936 is an available syn- onym [Article 60(b)]. Therefore, Buthoscorpio Werner 1936 = Pocockius Francke 1985, NEW SYNONYMY. This genus includes two species: Buthoscorpio politus (Pocock 1899), NEW COMBINATION (the type species) and B. sarasinorum (Karsch 1892), NEW COM- BINATION. Genus Compsobuthus Vachon 1949 Two taxa, originally described as Buthus acutecarinatus judaicus Birula 1905 (Middle East; type locality: Jordan and Lebanon) and Buthus acutecarinatus werneri Birula 1908 (Africa, Middle East; type locality: Sudan), were treated for a long time as two separate species of the genus Compsobuthus Vachon 1949. Levy & Amitai (1980) demonstrated that these two forms are subspecies of the same species (with an intergradation zone in Israel), assigned both subspecies to Compso- buthus werneri as C. werneri werneri and C. werneri judaicus. Recently, Sissom (1994) confirmed and redescribed C. werneri wer- neri. It can be observed that Buthus acutecari- natus judaicus Birula 1905 is a senior syn- onym of Buthus acutecarinatus werneri Birula 1908. The situation is complicated by the fact that Buthus judaicus Simon 1872 (currently Hottentotta judaicus) is found to be a primary senior homonym of Buthus acutecarinatus ju- daicus Birula 1905, NEW HOMONYMY (Article 53 of the Code). Therefore, the name Buthus acutecarinatus judaicus Birula 1905 is permanently invalid (Article 52b) and has to be replaced by the next available junior syn- onym which is Buthus acutecarinatus werneri Birula 1908 (currently Compsobuthus wer- neri). At the same time, the subspecies inhab- iting Israel, Jordan, and Lebanon should be called Compsobuthus werneri schmiedeknech- ti Vachon 1949 which is the next available junior synonym based on this population (type locality: Nazareth, Israel; originally described as Compsobuthus schmiedeknechti Vachon 1949). Genus Leiurus Ehrenberg 1828 The description of subspecies Androctonus {Liurus) quinquestriatus aculeatus Ehrenberg 1831 (as ""forma a aculeata"') was based on the same type specimens and locality (Egypt and Sudan) as that of Androctonus {Leiurus) quinquestriatus Ehrenberg 1828 (currently Leiurus quinquestriatus). Therefore, Androc- tonus {Leiurus) quinquestriatus Ehrenberg 1828 = Androctonus {Liurus) quinquestriatus aculeatus Ehrenberg 1831, NEW SYNONY- MY. Genus Lychas C.L. Koch 1845 C.L. Koch (1837) described the genus Pil- umnus without designating a type species or listing any species under this name. Later he (C.L. Koch 1850) found this name to be a junior homonym of Pilumnus Leach 1815 (Crustacea) and proposed to use instead the name Lychas. This was not, however, a new replacement name, since C.L. Koch (1845) described four species listed under the genus Lychas but did not give a separate description of this genus. This, according to Article 12(b)(5), constitutes an indication; therefore the correct date of the generic name Lychas is 1845, as correctly suggested by Vachon (1985) but not 1850, as used by Kraepelin (1899) and L.E. Koch (1977). Of the four species described in 1845 (and listed also in C.L. Koch 1850), only one, Ly- chas scutilus C.L. Koch 1845, currently is in- cluded in genus Lychas; all three other species are synonyms of Isometrus maculatus (De Geer 1778). On this basis, Pocock (1899) FET— OLD WORLD SCORPIONS 247 fixed the type species for the genus Lychas C.L. Koch 1845 as Lychas scutilus C.L. Koch 1845. I designate here the type species for Pil~ umnus C.L. Koch 1837, also as Lychas scu- tilus C.L. Koch 1845, which follows the re- quirements of Article 13(a). Francke (1985) introduced Repucha as a re- placement name for Pilumnus C.L. Koch 1837, and synonymized it with Lychas. How- ever, since the type species of Pilumnus C.L. Koch was not originally designated, Repucha Francke also does not have a type species and therefore (being created after 1930) is not available under Francke’s authorship [Article 13(b)]. I designate here the type species for Repucha as Lychas scutilus C.L. Koch 1845, which makes this generic name available as Repucha Fet 1997, and a junior synonym of Lychas C.L. Koch 1845. Lychas marmoreus obscurus Kraepelin 1916, NEW HOMONYMY, from Australia, is found to be a primary junior homonym (Ar- ticle 57) of Lychas asper obscurus Kraepelin 1913 from Africa. A new replacement name is introduced, Lychas marmoreus lucienkochi Fet, NOMEN NOVUM = Lychas marmoreus obscurus Kraepelin 1916, NEW SYNONY- MY. Etymology: a patronym honoring Dr. Lu- cien E. Koch, the author of a revision of Aus- tralian scorpiofauna (L.E. Koch 1977); a composite word is constructed to avoid con- fusion with the names of two other prominent scorpion taxonomists, Carl L. Koch and his son Ludwig Koch. This subspecies is listed by L.E. Koch (1977) as a valid form. Genus Pectinibuthus Fet 1984 Orlov & Vasilyev (1984) published Fet’s description of the new genus {Pectinibuthus) and its only species (as "'Pectinibuthus birulai Fet 1983”) from Turkmenistan without the au- thor’s permission, without information on type material, and with several mistakes. Fet (1989) treated these names as nomina nuda, since the extended correct description with the information on type material was published separately (Fet 1987). However, the 1984 date satisfies all requirements for the publication, availability of both genus and species names (Articles 11, 13), and fixation of the type spe- cies by indication (by monotypy) [Article 68(d)]. It appeared in a numerous-copy bro- chure (Orlov & Vasilyev 1984) published by the Gorky State University (Gorky, USSR); 300 copies of it were simultaneuosly obtain- able free of charge. The brochure is dated 1983 on the cover but was approved for print only in January 1984 (information on the back side of the cover page). Therefore, the correct date of publication for Pectinibuthus Fet and P. birulai Fet is 1984, and the correct refer- ence to the original description is “Fet in Or- lov et Vasilyev 1984”. Genus Uroplectes Peters 1862 Uroplectes fischeri intermedius Caporiacco 1941, NEW HOMONYMY, is found to be a primary junior homonym (Article 57) of Uro- plectes intermedius Tullgren 1907 (which is a junior synonym of Uroplectes xanthogram- mus Pocock 1897) (both from Africa). A new replacement name is introduced, Uroplectes fischeri caporiaccoi Fet, NOMEN NOVUM = Uroplectes fischeri intermedius Caporiacco 1941, NEW SYNONYMY Etymology: a pa- tronym honoring Dr. Lodovico di Caporiacco, the well-known Italian arachnologist. FAMILY CHACTIDAE POCOCK 1893 Genus Euscorpius Thorell 1876 Hadzi (1929) studied three European spe- cies, E. italicus (Herbst 1800), E. carpathicus (Linnaeus 1767), and E. germanus (C.L. Koch 1837) from the former Yugoslavia and adja- cent areas. Within each of those species, Had- zi described three “forms” which were given names oligotrichus, mesotrichus and polytri- chus. These forms have status of subspecies [Article 45(f)]. Capra (1939) correctly recog- nized E. italicus mesotrichus Hadzi 1929 as a (primary) senior homonym of E. germanus mesotrichus Hadzi 1929 and E. carpathicus mesotrichus Hadzi 1929 [Article 52(a)]. Capra did not introduce any replacement names. None of the three subspecies described by Hadzi (1929) within E. italicus is currently recognized as valid; however, these names re- main available junior synonyms of E. italicus. All six subspecies described by Hadzi (1929) within E. germanus and E. carpathicus are primary junior homonyms. Most of these forms do not have originally designated type specimens and/or localities. Type material of Hadzi, formerly in the Slovenian Academy of Sciences in Ljubljana, is considered lost (M. Kuntner pers. comm.). For E. carpathicus po- ly trichus Hadzi 1929 (type locality unknown), Caporiacco (1950) published a replacement 248 THE JOURNAL OF ARACHNOLOGY name, E. carpathicus hadzii. The following observations can be made regarding the re- maining five subspecies. E. carpathicus oligotrichus Hadzi 1929, NEW HOMONY- MY (type locality unknown) and E. germanus polytrichus Hadzi 1929, NEW HOMONYMY (type locality unknown) are not diagnosable at the subspecies level and are both synonyms of Euscorpius carpathicus (L. 1767) (Capo- riacco 1950). E. germanus oligotrichus Hadzi 1929, NEW HOMONYMY (type locality unknown) and E. germanus mesotrichus Hadzi 1929 (type locality: Kranjska, now Slovenia) are also not diagnosable (Caporiacco 1950). Both forms were synonymized with E. germanus by Kinzelbach (1975). According to current di- vision (Bonacina 1980), they may belong ei- ther to E. germanus (C.L. Koch 1837) or to E. mingrelicus (Kessler 1874). The validity and rank of Euscorpius car- pathicus mesotrichus Hadzi 1929 (type local- ity: southern Slovenia) remains unclear. Ca- poriacco (1950) synonymized it with E. carpathicus tergestinus (C.L. Koch 1837) (type locality: Trieste, Italy). Kinzelbach (1975) did not accept this synonymy and el- evated Hadzi ’s subspecies to the species status as Euscorpius mesotrichus Hadzi 1929, sig- nificantly increasing its scope and range. A number of authors (e.g., Michalis & Dolkeras 1989; Lacroix 1991; Kritscher 1993) followed Kinzelbach (1975) in using the name E. me- sotrichus, although it is a primary junior hom- onym and cannot be used. If this form is con- sidered a valid species, it currently should be called Euscorpius tergestinus (C.L. Koch). The subspecies Euscorpius germanus gam- ma Caporiacco 1950 was based on a series of syntypes from northeastern Italy and Slovenia. This subspecies was revised by Bonacina (1980) who synonymized part of it with the nominotypical subspecies E. germanus ger- manus (C.L. Koch), while transferring another part as a subspecies to Euscorpius mingrelicus (Kessler). For this latter subspecies, Bonacina (1980) introduced a replacement name, E. mingrelicus caprai. However, Caporiacco’s name remains available even if it denotes more than one taxon (Article 17). Therefore, the correct name for this subspecies is Eus- corpius mingrelicus gamma (Caporiacco 1950) (in part), NEW COMBINATION = Euscorpius mingrelicus caprai Bonacina 1980, NEW SYNONYMY. FAMILY ISCHNURIDAE SIMON 1879 Genus Hadogenes Kraepelin 1894 Hadogenes trichiurus paucidens Werner 1939, NEW HOMONYMY, is a primary ju- nior homonym of Hadogenes paucidens Po- cock 1896 (both from South Africa). A new replacement name is introduced, Hadogenes trichiurus werneri Fet, NOMEN NOVUM — Hadogenes trichiurus paucidens Werner 1939, NEW SYNONYMY. Etymology: a patronym honoring Dr. Franz Werner who made exten- sive contributions to scorpion taxonomy in the 1900s-1930s. The validity of this form was never challenged probably because it was for- gotten; it was neither listed by Lamoral & Reynders (1975) nor discussed in the recent revision of Hadogenes (Newlands & Cantrell 1986). FAMILY SCORPIONIDAE LATREILLE 1802 Genus Pandinus Thorell 1876 Two taxa, Pandinoides Vachon 1974 and Pandinurus Vachon 1974, were described (Vachon 1974) as subgenera of Pandinus; however, type species were not designated or indicated for these (non-monotypic) taxa. Ac- cording to Article 13(b) of the Code, these names are not available under Vachon’s au- thorship. I designate here their type species and retain the generic names (as described by Vachon 1974): Pandinoides Fet 1997 (type species Scorpio exitialis Pocock 1888); and Pandinurus Fet 1997 (type species Pandinus militaris Pocock 1900). Three other valid subgenera of Pandinus are: the nominotypical subgenus (type species by original designation Buthus imperator C.L. Koch 1841), Pandinops Bimla 1913 (type species by indication Pandinus peeli Pocock 1900), and Pandinopsis Vachon 1974 (type species by monotypy Scorpio dictator Pocock 1888). Genus Scorpio Linnaeus 1758 The species Scorpio maurus subtypicus Bi- rula 1910, NEW HOMONYMY (from Mo- rocco) is found to be a primary junior hom- onym of Scorpio africanus subtypicus Kraepelin 1894 (from Sudan; currently Pan- FET— OLD WORLD SCORPIONS 249 dinus imperator subtypicus), and therefore is a permanently invalid name (Article 52 of the Code). A new replacement name is intro- duced, Scorpio maurus birulai Fet, NOMEN NOVUM ™ Scorpio maurus subtypicus Bimla 1910, NEW SYNONYMY. Etymology: a pa- tronym honoring the famous Russian scor- piologist Dr. Alexei A. Birala (Byalynitsky- Bimla). ACKNOWLEDGMENTS I am grateful to Norman Plateick (Ameri- can Museum of Natural History, New York), W. David Sissom (West Texas A & M Uni- versity, Canyon, Texas), Graeme Lowe (Mo- nell Chemical Senses Center, Philadelphia), and two anonymous reviewers for their com- ments and help in taxonomic questions. Philip Tubbs (International Commission on Zoolog- ical Nomenclature, London) Medly advised on the Code matters. Matjaz Kuntner (Ljubljana, Slovenia) helped to clarify information on taxa decribed by J. Hadzi from former Yu- gO'Slavia. Mark Volkovich (Zoological Insti- tute, Russian Academy of Sciences, St. Pe- tersburg) kindly provided photocopies of rare publications by Birala and Hadzi. I also thank Matt E. Braunwalder (Zurich, Switzerland) for his invaluable help with bibliographic re- search. LITERATURE CITED Birula, A. 1905. Skorpiologische Beitrage. Zool. Anz., 29:621-624. Birula, A. 1908. Ergebnisse der mit Subvention aus der Erbschaft Treitl untemommenen zoolo- gischen Forschungsreise Dr. F. Werner's nach dem Anglo-Aegyptischen Sudan und Nord- Uganda. XIV. Skorpiones und Solifugae. Sit- zungsb. Kais.-Konigl. Akad. Wiss., Wien, 117: 12U152. Bonacina, A. 1980. Sistematica specifica e sottos- pecifica del complesso '‘Euscorpius germanus'' (Scorpiones, Chactidae). Riv. Mus, Civ. Sci. Na- tur. “EnricO' Caffi” (Bergamo), 2:47-100, Caporiacco, L. di. 1950. Le specie e sottospecie del genre “Euscorpius” viventi in Italia ed in alcune zone confinanti. Mem. Accad. Lincei, CL Sci. Fis. Mat. Nat. (8) 2, 3a, 4:159-230. Capra, F. 1939. llEuscorpius germanus (C.L. Koch) in Italia (Arachnida, Scorpiones). Mem, Soc. En- tomol. ItaL, 18:199-213. Fet, V. 1987. A new genus and species of a scor- pion from East Karakum: Pectinibuthus birulai Fet n. g. n. sp. (Scorpiones, Buthidae). Entomol. Obozr. (Rev. Entomol. USSR), 66:443-446 (in Russian, English summary). Fet, V. 1989. A catalogue of scorpions (Chelicer- ata: Scorpiones) of the USSR. Riv. Mus. Civ, Sci. Natur. “Enrico Caffi” (Bergamo), 13:73- 171. Francke, O.E 1985. Conspectus genericus scor- pionoram 1758-1982 (Arachnida: Scorpiones). Occ. Papers Texas Tech Univ., 98:1-32. Hadzi, J. 1929. Die Skorpione der Schmidt'schen Saminlung (Euscorpius italicus polytrichus n. ssp. und andere neue Rassen). Glasn. Muz. Drast. Slovenijo, (B), 10:30-41. ICZN. 1985. International Code of Zoological No- menclature, 3rd ed. Univ. of California Press, Berkeley. Karsch, F. 1892, Arachniden von Ceylon und von Minikoy gesammelt von den Herren Doctoren P. und E Sarasin. Berliner Entomol. Zeitschr., 36: 267-310. Kinzelbach, R. 1975. Die Skorpione der Aegais. Beitrage zur Systematik, Phylogenie und Biogeo- graphie. Zool. Jahrb., Abt. Syst., 102:12-50. Koch, C.L. 1837. Scorpionen. In Uebersicht des Arachnidensy stems. C.H. Zeh’sche Buchhan- dlung, Ntimberg, 1:1-39. Koch, C.L. 1845. Die Arachniden. C.H. Zeh’sche Buchhandlung, Ntimberg. 12:1-166. Koch, C.L. 1850. Scorpionen. In Uebersicht des Arachnidensy stems. C.H. Zeh’sche Buchhan- dlung, Ntimberg, 5:86-92. Koch, L.E. 1977. The taxonomy, geographic dis- tribution and evolutionary radiation of Australo- Papuan scorpions. Rec. Western Australian Mus., 5:83-367. Kraepelin, K. 1899. Scorpiones und Pedipalpi, In Das Tierreich (F. Dahl, ed.). R, Friedlander und Sohn Verlag, Berlin, 8:1-265. Kritscher, E. 1993. Ein Beitrag zur Verbreitung der Skorpione im ostlichen Mittelmeerraum. Ann. Naturh. Mus. Wien (B), 94/95:377-391. Lacroix, J.-B. 1991. Faune de France; Arachnida: Scorpionida. 5e note. Sub-genus (Euscorpius) Thorell, 1876. Arachnides, 8:17-36. Lamoral, B.H. 1979. The scorpions of Namibia (Arachnida: Scorpionida). Ann. Natal Mus., 23: 497-784. Lamoral, B.H. & S.C. Reynders. 1975. A catalogue of the scorpions described from the Ethiopian faunal region up to December 1973. Ann. Natal Mus., 22:489-576. Levy, G. & P. Amitai. 1980. Scorpiones. Fauna Pa- laestina. Arachnida I. The Israel Acad. Sci. Hu- man., Jerusalem. 130 pp. Michalis, K. & P. Dolkeras. 1989. Beitrag zur Kenntnis der Skorpione Thessaliens und Epirus (Nordgriechenland), Entomol. Mitt. Zool. Mus. Hamburg, 9:259-270. Newlands, G. & A.C. Cantrell, 1986. A re-apprais- 250 THE JOURNAL OF ARACHNOLOGY al of the rock scorpions (Scorpionidae: Hado- genes). Koedoe, 28:35-45. Orlov, B.N. & N.F. Vasilyev. 1984. Scorpions and their venom. Part 2. The key to the scorpions of the USSR fauna; ontogenesis; conservation. Gor- ky State University, Gorky, 32 pp. (in Russian). Pocock, R.I. 1899. On the scorpions, pedipalps and spiders from Tropical West Africa represented in the collection of the British Museum. Proc. Zool. Soc. London, 1899:833-885. Sissom, W.D. 1994. Descriptions of new and poor- ly known scorpions of Yemen (Scorpiones: Buth- idae, Diplocentridae, Scorpionidae). Fauna of Sa- udi Arabia, 14:3-39. Vachon, M. 1941. Sur un Scorpion presaharien type d’un nouveau genre Trichobuthus grubleri n. sp. Bull. Soc. Zool. France, 66:339-350. Vachon, M. 1942. Remarques sur un Scorpion pre- desertique peu connu Buthiscus bicalcaratus Bi- rula. Bull. Mus. Natnl. Hist. Natur. (Paris), (2)14: 419-421. Vachon, M. 1952. Etude sur les Scorpions. Institut Pasteur d’Algerie, Alger, 482 pp. Vachon, M. 1961. A propos d’un Scorpion de ITnde: Buthoscorpio laevicauda Werner (famille des Scorpionidae), synonyme de Stenochirus pol- itus Pocock, 1899 (famille des Buthidae). Bull. Soc. Zool. France, 86:789-795. Vachon, M. 1974. 6tude des caracteres utilises pour classer les families et les genres de Scor- pions (Arachnides). 1. La trichobothriotaxie en Arachnologie. Sigles trichobothriaux et types de trichobothriotaxie chez les Scorpions. Bull. Mus. Natnl. Hist. Natur. (Paris), (3)140 (Zool. 104): 857-958. Vachon, M. 1985. Contributions a la revision du genre Lychas C.L. Koch, 1845 (Arachnida, Scor- piones, Buthidae) 1. Historique du nom gene- rique. Rev. Arachnol., 6:99-106. Manuscript received 30 August 1996, accepted 6 May 1997. 1997. The Journal of Arachnology 25:251-256 DESCRIPTION OF THE MALE OF DIPLOCENTRUS LOURENCOI (SCORPIONES, DIPLOCENTRIDAE) Kari J. McWest: Department of Life, Earth, and Environmental Sciences; West Texas A&M University, WTAMU Box 808, Canyon, Texas 79016 USA ABSTRACT. The male of Diplocentrus lourencoi Stockwell 1988 is described and illustrated from a specimen collected west of San Pedro Sula, Honduras, and compared to the holotype female and the males of other Diplocentrus Peters 1861 in the region. The hemispermatophore is described, illustrated, and compared to other Diplocentrus in the region. Unique new characters on the male, posterolateral recesses on tergite VII, are described and illustrated. New descriptive information is given for the holotype female. Complete measurements and morphometries are given for the male and female. A comparative diagnosis is offered based on this new information. The original description of Diplocentrus lourencoi Stockwell 1988 was based on a sin- gle adult female from Rio Santa Ana Canyon (3500 ft.), San Pedro Sula, Departamento Cor- tes, Honduras, collected during the Field Mu- seum Expedition into Central America in the spring of 1923. Recently, Thomas G. Anton of the Field Museum (FMNH) discovered un- determined scorpion material that included an additional specimen from the expedition that was not studied previously. This specimen is an adult male collected by K. Schmidt and L. Walters on 1 April 1923 at “Mt. Camp, 4500 ft. EL, W. of San Pedro Sula, Honduras,” sev- eral days after the female holotype was col- lected. (According to Mr. Anton [pers, comm.], “Mt. Camp” refers to the campsite location and not a geographic place name.) Examination of the holotype female was nec- essary to confirm that this male specimen is referable to D. lourencoi. This specimen brings the total number of reported diplocen- trids from Honduras to a mere 12 individuals assignable to four species in two genera: Di- plocentrus coddingtoni Stockwell 1988, D. santiagoi Stockwell 1988, D. lourencoi, and Didymocentrus krausi Francke 1978. Because the male of this species is previ- ously undescribed and distinct sexual dimor- phism in diplocentrids is well documented, it is important that comparisons are made to complete the diagnosis of the species and to aid in separating D. lourencoi from related species. Male morphology tends to provide more diagnostic characters for the genus at the species level than do female characters (Francke 1977). Due to the brevity of the original descrip- tion it became necessary to examine the fe- male holotype. Upon exanfination of the ho- lotype, the drawings of the female pedipalp chela in the original description were found to depict inaccurately the nature of the chelal re- ticulation, granulation, and keel structure. The pattern of chela reticulation and texture ob- served in the holotype female is extremely similar to that seen in the male, which is here- in described and illustrated. In addition, the notable slenderness of the chelicerae and length of the cheliceral fingers of Diplocen- trus lourencoi were not included in the orig- inal description: cheliceral morphometries have been found to have diagnostic impor- tance in diplocentrid scorpions (Francke 1977). Complete measurements for the holo- type female are published here for the first time. Nomenclature and mensuration essentially follow that of Stahnke (1970), except that cheliceral measurements and carinal terminol- ogy are after Francke (1975, 1977) and trich- obothrial terminology is after Vachon (1974). Diplocentrus lourencoi Stockwell 1988 Diplocentrus lourencoi Stockwell 1988: 161-163, figs. 19-24. Type data. — Holotype female from Rio Santa Ana Canyon (3500 ft.), San Pedro Sula, 251 252 THE JOURNAL OF ARACHNOLOGY Departamento Cortes, Honduras, 21 March 1923 (K. Schmidt and L. Walters), Capt. Field Mus. Exped., deposited in the Field Museum, Chicago; examined {Note: The original data label with the specimen does not indicate San Pedro Sula or Departamento Cortes: this in- formation was thoughtfully provided in the original description by Stockwell, apparently after a thorough investigation of Honduras ge- ography). Description of male. — Coloration (in al- cohol): Carapace dark orange brown to ma- hogany with weak underlying marbling of light patches. Tergites dark orange brown to mahogany, infuscate. Metasoma dark orange brown, with each segment gradually darker distally; dorsal keels at least proximally ni- grocarinate; telson dark orange brown, grad- ing to light yellow brown at subaculear tubercle. Venter of prosoma reddish brown be- coming yellow brown laterally with faint in- fuscation; genital opercula, pectines, and basal piece yellowish. Stemites II-VI light orange brown to light yellow brown; stemite VII slightly darker. Chelicera manus light yellow brown becoming slightly darker distally with faint reticulate pattern proximally; fingers yel- lowish, with teeth somewhat transparent under magnification, appearing light orange brown. Pedipalp external surfaces deep mahogany gradually becoming lighter distally to light red-brown fingers; internal surfaces light yel- low brown at dorsal lobe becoming darker dis- tally. Legs dark yellow brown proximally with marbled pattern, distal segments light yellow- ish. Prosoma: Carapace only slightly longer than posterior width; length/width ratio 1.05. Surface feebly reticulate. Anterior margin of carapace and anterior median furrow moder- ately coarsely granular (Fig. 1); remainder of carapace with dense fine granulation inter- spersed with coarser granules. Sternum with 14 setae; each genital operculum with seven setae. Coxosternal region finely punctate, somewhat lustrous; coxae sparsely setose. Mesosoma: Tergites I-VI with very dense fine granulation, shagreened. Tergite VII weakly bilobed posteriorly with moderate to coarse granulation (Fig. 2). Pre-posterolateral recess moderate, shallow; posterolateral recess strong, deep (Fig. 2). Pectinal tooth count 9- 10. Stemites III to VI finely punctate and somewhat lustrous, smooth. Stemite VII with submedian carinae vestigial, smooth, provided with four pairs of setae; lateral carinae weak, smooth, provided with five pairs of setae. Hemispermatophore: Lamelliform; inner margin of median lobe with a moderate ridge with four teeth. Distal lamina not broad, ta- pering only at distal end (Figures 3-5). Metasoma: Segments I-IV: Dorsolateral ca- rinae moderate, granular, increasing in strength on distal segments. Lateral suprame- dian carinae on I-IV strong, on I-II smooth to granular, on III-IV granular throughout. Lateral inframedian carinae on I-III moderate, smooth to granular; on IV (Fig. 6) represented by an irregular row of granules. Ventrolateral carinae on I moderate, smooth; on II-IV weak, granular. Ventral submedian carinae on I and II weak, smooth to granular; on III mod- erate, smooth on proximal one-half, vestigial with irregularly scattered granules on distal half; reduced on IV to two narrowly separated irregular rows of granules on proximal two- thirds. Intercarinal spaces feebly reticulate. Segment V: Dorsal surface densely finely granular medially with coarser granulation lat- erally; dorsolateral carinae moderate, densely moderately granular; lateromedian carinae present only on anterior two-thirds, strong, granular; lateral intercarinal surfaces feebly reticulate, smooth distally (Fig. 6); ventrolat- eral, ventromedian, and ventral transverse ca- rinae moderate to strong with subconical gran- ules, with cluster of seven ventromedian granules beyond ventral transverse row. Telson: Ventral surface hirsute with short, white microchaetes outnumbering light-red setae; densely covered with extremely fine granules, increasingly coarser ventrally and proximally (Fig. 6). Subaculear tubercle densely hirsute, strong, laterally compressed, subconical in profile, with three pairs of long, light-red setae. Chelicerae: Dentition as in Fig. 7. Smooth and relatively long, slender; fixed finger short- er than chela width (0.82X); movable finger subequal to chela length (0.93 X). Other ratios in morphometric ratios section. Ventral brush- es thick, long. Teeth largely transparent with only slight coloration on di of movable finger; di and de not in apposition (Fig. 8). Pedipalps: Trichobothria pattern Type C, orthobothriotaxic (Vachon 1974). Femur: Wider than deep, with dorsointemal and ven- trointemal carinae strong, granulose; dorsoex- McWEST—DIPLOCENTRUS LOURENCOI MALE DESCRIPTION 253 Figures 1-8. — Morphology of male of Diplocentrus lourencoi Stockwell. 1, Dorsal aspect of anterior region of carapace; 2, Oblique view of right lateral aspect of tergite VII, showing posterolateral recesses, ppr — pre-posterolateral recess; pr = posterolateral recess; 3, Dorsal aspect of right hemispermatophore; 4, Detail of inner margin of median lobe, showing positions of teeth; 5, Ventral aspect of right hemi- spermatophore; 6, Left lateral aspect of metasomal segments IV, V, and telson; 7, Dorsal aspect of left chelicera; 8, External aspect of left cheliceral movable finger. temal carina strong, irregularly granule se on basal two-thirds, moderately granulose on dis- tal third; ventroextemal carina vestigial to ob- solete, moderately granulose at proximal one- fifth. Dorsal surface of femur flat, with mod- erate and coarse granules scattered mostly for % of length, dorsomedian seta situated cen- trally (Fig. 9). Internal face covered with coarse granules, several of these granules rath- er large. Ventral surface finely granulose ex- ternally, gradually becoming coarsely granular toward internal face. Patella: Dorsomedian keel moderate, smooth; dorsoextemal keel moderate, irregular, smooth to granular (Fig. 10); ventrointemal keel moderate, smooth to coarsely granular; ventromedian keel reduced, with scattered fine to moderate granules; ven- troextemal keel moderate, smooth, distal third comprised of strong reticular costae. Chela: Palm (Fig. 11) with distinct reticulate pattern formed by weak, unpigmented costae; costae lustrous, interspersed with moderate to coarse 254 THE JOURNAL OF ARACHNOLOGY Figures 9-12. — Morphology of the pedipalps of Diplocentrus lourencoi. 9, Dorsal aspect of right ped- ipalp femur, dms = dorsomedian seta; 10, Dorsal aspect of right pedipalp patella; 11, External aspect of right pedipalp chela; 12, Dorsal aspect of right pedipalp chela. granules; intercostal areas extremely finely granular (shagreened); dorsal margin (Fig. 12) moderately granular basally to strongly, coarsely granular to base of fixed finger; dig- ital keel moderate and smooth for most of length, granulose and fading at trichobothrium Dt\ dorsal and external secondary keels re- duced, wide, and moderately granulose; outer surface above imaginary line between tricho- bothria Esb and Est reticulate, below line granulose; ventral keel strong, coarsely gran- ular; ventrointemal keel moderate, granulose; internal surface of palm smooth, feebly retic- ulate with all keels greatly reduced and inter- spersed with fine to moderate granules. Fixed and movable fingers punctate, setose. Legs: Tarsomere II spine formula 4/5 4/5: 5/5 5/5: 5/5 5/5: ?/? 5/5. Measurements: Total L, 53.8; carapace L/W, 6. 5/6. 2; mesosoma L (specimen fragile and longitudinally compacted), 14.5; metaso- ma L, 26.9; telson L, 5.9. Metasomal seg- ments: I L/W, 4. 1/3.8; II L/W, 4.7/4.0; III L/W, 5.0/3.4; IV L/W, 5.7/3.0; V L/W, 7.4/2.7. Tel- son: vesicle L/W/D, 4.4/2. 8/2.4; aculeus L, 1.5. Pedipalps: Femur L/W, 6. 2/2. 2; patella L/W, 6.6/1.2; chela L/W/D, 12.8/3.2/5.0; fixed finger L, 6.2; movable finger L, 8.4; palm (un- derhand) L, 4.4. Chelicerae: Manus L/W, 2.01/ 1.35; fixed finger L, 1.11; movable finger length (Hide: 1.86/1.47. Comparison with holotype female. — ^The male is similar in appearance to the female with notable exceptions. Following are the morphometric ratios, with female ratios in pa- rentheses. Pedipalp proportionately longer and more slender, pedipalp femur length/width, 2.82 (2.35); patella length/width, 3.14 (2.20); chela length/width, 4.00 (3.43); length/depth, 2.56 (2.26); metasomal segment II slightly shorter, length/width, 1.18 (1.22); metasomal segment V considerably longer, length/width, 2.74 (2.38). (Other ratios provided below.) Pedipalps: Granules noticeably more reduced, especially on dorsal surfaces of femur and chela, rarely encountered in diplocentrid sys- McWEST—DIPLOCENTRUS LOURENCOI MALE DESCRIPTION 255 tematics but known to occur in Diplocentrus rectimanus Pocock (Francke 1977). Dorsal surface of femur flat on male, distinctly con- vex on female. Reticulation of chela less dis- tinct. Telson distinctly less granulose. Female slightly lighter in color, being a more yellow- ish brown than reddish brown, keeping in mind that the specimens are old and that some integumental separation has occurred in the female. Integument of carapace and tergites shagreened in male, somewhat lustrous in fe- male. Tergite VII: The female has a more typi- cally shaped disc. The male, on the other hand, has strong, deep, posterolateral recesses (Fig. 2). Upon further comparisons with sev- eral species of Diplocentrus, Bioculus Stanke 1968, Didymocentrus Kraepelin 1905, and Nebo Simon 1878 (made available by W. Da- vid Sissom and Chad M. Lee), it became ev- ident that the extreme depth of this feature is unique to at least this male specimen. Unfor- tunately, only a single male specimen of D. lourencoi is known so the utility of this fea- ture in diplocentrid systematics will require confirmation as new material accumulates. Future descriptions should, therefore, include a brief statement regarding this feature. Measurements of holo type female: Total L, 50.9; carapace LAV, 6. 3/6.4; mesosoma (dis- tended) L, 16.6; metasoma L, 22.3; telson L, 5.7. Metasomal segments: I L/W, 3. 5/3.4; II L/W, 3.9/3.2; III L/W, 4. 1/3.1; IV L/W, 4.6/ 2.9; V L/W, 6.2/2.6. Telson: Vesicle L/W/D, 4.4/2. 8/2. 3; aculeus L, 1.3. Pedipalps: Femur L/W, 5.4/2.3; patella L/W, 5.5/1.7; chela L/W/ D, 12.0/3.5/5.3; fixed finger L, 5.1; movable finger L, 7.4; palm (underhand) L, 4.6. Che- licerae: Manus L/W, 1.95/1.47; fixed finger L, 1.11; movable finger length dUde: 1.89/1.53. Morphometric ratios: (Female ratios are in parentheses.) Carapace L/W 1.05 (0.98); metasoma II L/W 1.18 (1.22); metasoma III L/W 1.47 (1.32); metasoma V L/W 2.74 (2.39); chelicera chela L/W 1.49 (1.33), fixed finger L/chela W ratio 0.82 (0.76), movable finger L/chela L ratio 0.93 (0.97); pedipalp fe- mur W/D (at dorsomedian seta) 1.22 (1.15), chela fixed finger L/carapace L 0.95 (0.81), movable finger L/carapace L 1.29 (1.17), che- la L/W 4.00 (3.43), chela L/D 2.56 (2.26), chela W/D 0.64 (0.66), movable finger L/che- la D 1.68 (1.40), movable finger L/metasoma V L 1.14 (1.19), fixed finger L/carapace L ra- tio 0.95 (0.81). Comparative diagnosis. — The female of Diplocentrus lourencoi was compared to other Diplocentrus in the region by Stockwell 1988. Comparisons of the male with other species are based on the original descriptions and il- lustrations (i.e., the specimens were not ex- amined). The hemispermatophore of D. lour- encoi does not bear spines on the anterior margin of the median lobe as do those of D. steeleae and D. ornatus. The hemispermato- phore of D. coddingtoni differs by lacking denticles on the inner margin of the median lobe. The pedipalp chela of D. lourencoi dif- fers from that of D. omatus by its greater width/depth ratio (0.64 versus 0.43) and great- er length/depth ratio (2.56 versus 2.16); it differs from D. coddingtoni by its longer, more robust chela (length/depth ratio 2.56 ver- sus 2.98; chela length/fixed finger length ratio 2.04 versus 2.44). ACKNOWLEDGMENTS I wish to thank my good friend Tom G. An- ton (formerly an assistant at EMNH) for no- ticing undetermined scorpion material in the EMNH collection and subsequently reporting such specimens to W. David Sissom and me. His shared knowledge of the 1923 herpetolog- ical Field Museum Expedition into Central America was also very helpful. Thanks are also due to A1 Newton for granting loans of this material and to Daniel Summers and Phil- ip P. Parrillo (EMNH) for preparing the loans of such material and the holotype female of Diplocentrus lourencoi. I also wish to thank Chad M. Lee for discussions on diplocentrids and for his reviews of the working drafts of the manuscript. William M. Burrell of the Dept, of Fine Arts at Amarillo College, Am- arillo, Texas is greatly appreciated for his ar- tistic suggestions and support. Special thanks are reserved for Dr. Sissom for obtaining the specimens and passing them along to me for examination. He also allowed me to peruse his private collection and other specimens in his care. I am extremely grateful for his guidance through the course of this study and for his suggestions and criticism while reviewing the various drafts of the manuscript. Petra Sier- wald graciously furnished valuable sugges- tions incorporated into the manuscript. Victor Fet and Emilio Maury kindly reviewed the 256 THE JOURNAL OF ARACHNOLOGY manuscript. This study was supported with a West Texas A&M University Research Assis- tantship. LITERATURE CITED Francke, O.E 1975. A new species of Diplocentrus from New Mexico and Arizona (Scorpionida, Di- plocentridae). J. ArachnoL, 2:107-118. Francke, O.E 1977. Scorpions of the genus Diplo- centrus from Oaxaca, Mexico (Scorpionida, Di- plocentridae). J. ArachnoL, 4:145-200. Stahnke, H.L. 1970. Scorpion nomenclature and mensuration. Entomol. News, 81:297-316. Stockwell, S.A. 1988. Six new species of the genus Diplocentrus Peters from Central America (Scor- piones, Diplocentridae). J. ArachnoL, 16:153- 175. Vachon, M. 1974. Etude des caracteres utilises pour classer les families et les genres de Scor- pions. (Arachnides). Bull. Mus. Nat. Hist. Nat., Paris, 3nt smes. No. 140, ZooL, 104:857-958. Manuscript received 5 September 1996, revised 15 February 1997. 1997. The Journal of Arachnology 25:257-261 GUERROBUNUS VALLENSISy A NEW SPECIES OF HARVESTMAN (OPILIONES: PHALANGODIDAE), FROM A CAVE IN VALLE DE BRAVO, STATE OF MEXICO, MEXICO Ignacio M. Vazquez: Laboratorio de Acarologia, Facultad de Ciencias, Universidad Nacional Autonoma de Mexico, Coyoacan 04510 D.F., Mexico James C. Cokendolpher: Adjunct Professor, Department of Biology, Midwestern State University, Wichita Falls, Texas 76308 USA ABSTRACT. Caecoa Silhavy 1973 is syEonymized under Guerrobunus Goodnight & Goodnight 1945. The third species of Guerrobunus is named. A taxonomic key to the species of Guerrobunus is provided. Two males and one female of the new species Guerrobunus vallensis are illustrated and described from a cave in Valle de Bravo, State of Mexico, Mexico. RESUMEN. Caecoa Silhavy 1973 se sinonimiza con Guerrobunus Goodnight & Goodnight 1945. Se le da nombre a la tercera especie de Guerrobunus. Se presenta una clave taxonomica para las especies del genero. Se describen dos machos y una hembra de Guerrobunus vallensis nueva especie de una cueva de Valle de Bravo, Estado de Mexico, Mexico. The arachnological fauna from Mexican caves is very rich and its study is in progress. Opilionids thus far reported from Mexican caves belong to the Neogoveidae, Cosmetidae, Phalaiigodidae, Nemastomatidae and Sclero- somatidae (= Gagrellidae). The Phalangodi- dae has the highest number of cave-adapted species world-wide (Rambla & Juberthie 1994) as well as in Mexico (Reddell 1981); seven described by Goodnight & Goodnight (1945, 1953, 1971, 1973), nine by Silhavy (1974, 1977) and one each by Pickard-Cam- bridge (1904) and Shear (1977). Seven of these phalangodids are true eyeless troglobi- tes: Troglostygnopsis anophtalma Silhavy 1973 and MexotrogUnus sbordonii Silhavy 1977 from Chiapas; Troglostygnopsis inops (Goodnight & Goodnight 1971) from Tamau- lipas; Hoplobunus apoalensis Goodnight & Goodnight 1973 and Neogovea mexasca Shear 1977 from Oaxaca; Hoplobunus planus Good- night & Goodnight 1973 from San Luis Po- tosi; and Caecoa arganoi Silhavy 1973 from Veracruz. During explorations of the caves in the State of Mexico, three phalangodids that have eyes with clear cornea and unpigmented retina were collected. Because these specimens re- semble Guerrobunus minutus Goodnight & Goodnight 1945 and Caecoa arganoi, a study was undertaken to determine the identity of the new specimens and the relationship of the two monotypic genera. Herein the new spec- imens are described as a new species and Cae- coa is synonymized under Guerrobunus. In 1945, Goodnight & Goodnight described the new genus Guerrobunus to contain their new species, minutus. Later, those same au- thors (1953) synonymized Guerrobunus (along with 14 other genera) under Cynortina Banks 1909. Realizing that Cynortina was preoccupied, Goodnight & Goodnight (1983) transferred the species of '"Cynortina"^ known from Costa Rica to the next oldest genus, Da~ pessus Roewer 1933. This action resulted in those authors newly synonymizing seven gen- era (formerly listed as synonyms of Cynorti- na) and left seven of the genera which they had synonymized in 1953 unplaced. At that time, they also revalidated Stygnoleptus Banks 257 258 THE JOURNAL OF ARACHNOLOGY 1914 and newly synonymized four other gen- era under Stygnoleptus. Stygnoleptus and three of these genera had previously (1953) been considered by them to be synonyms of Cy- nortina. Three genera (Azaca Goodnight & Good- night 1942, Ethobunus Chamberlin 1925, and Guerrobunus) synonymized under Cynortina in 1953 should have been listed as synonyms of Dapessus by Goodnight & Goodnight (1983), but they were not. Although Azaca and Ethobunus are known from Costa Rica and Panama, respectively; neither were men- tioned in the Goodnight & Goodnight (1983) publication on the phalangodids of Costa Rica and will have to await further study to deter- mine their true identities. Interestingly, the fe- male and only known specimen of Azaca lon- ga (Goodnight & Goodnight 1942) was collected on the same day, location, and by the same person as the two known specimens (both males) of Dapessus parallelus (Good- night & Goodnight 1942). The lack of a listing of Ethobunus with Dapessus was questioned in the manuscript review by Cokendolpher (7 November 1981) of the paper by Goodnight & Goodnight (1983), and therefore it can be assumed that they had changed their mind on the 1953 synonymy. Furthermore, if Ethobu- nus is a synonym of Dapessus, it is the older name and would require a shifting of all spe- cific names currently listed under Dapessus. The rediagnosis of Dapessus by Goodnight & Goodnight (1983) clearly excludes Guer- robunus because of the higher number of tar- sal elements in species of Dapessus. Further- more, the male and female genitalia differ greatly. Guerrobunus Goodnight & Goodnight Guerrobunus Goodnight & Goodnight 1945:1. Caecoa Silhavy 1974:189 (new synonymy). Diagnosis. — Small to medium sized phal- angodids, body length 1. 6-3.0 mm, cephalo- thorax narrower anteriorly; with several ob- tuse tubercles on anterolateral comers. With five distinct thoracic areas, first without a me- dian line, three free abdominal tergites. Body and leg surfaces without spines, smooth or with small granulations and tiny setae. Eye mound hemispherical, without armament, with or without eyes, at the anterior margin or slightly removed. Maxillary lobes of second coxae with ventral projections variable in size. Spiracles not visible. Tarsal segments: 3:4: 4(5) :5, both distitarsi I and II with two seg- ments. Penis with sclerotized tmncus, stylus and glans soft; tmncus with paired terminal ventral apophyses, sometimes also with dorsal pair of apophyses. Ovipositor short, with many setae and pair of apophyses at the distal end. Key to the Species of Guerrobunus la. Eyes absent, large ventral projections on maxillary lobes of coxae II present (§il- havy 1973, fig. 40) (from State of Mexico). G. arganoi lb. Eyes present (retina may be unpigment- ed), projections on maxillary lobes small (Fig. 9) ......................... . 2 2a. Retina of eyes darkly pigmented, eye mound with small rounded tubercles, male body length less than 1 .7 mm (from Guerrero) G. minutus 2b. Cornea clear, retina unpigmented, eye mound smooth, male body length more than 2.5 mm (from State of Mexico) G. vallensis new species Guerrobunus minutus Goodnight & Goodnight Guerrobunus minutus Goodnight & Goodnight 1945:1. Cynortina minutus: Goodnight & Goodnight 1953: 15 (by implication). Comments. — Examination of the female holotype (from American Museum of Natural History) revealed that the ovipositor had been removed and is apparently lost. The “female” paratype (Universidad Nacional Autonoma de Mexico) was also examined and determined to be a male. The penis was not illustrated or described because the curator of the museum did not allow the dissection. Goodnight & Goodnight (1977) described a new species, Cynortina minutus from Belize which was a secondary homonym of Cynor- tina (= Guerrobunus) minutus. As they are no longer considered to be congeneric they are no longer homonyms requiring a replacement name. Guerrobunus vallensis new species Figs. 1-10 Diagnosis and comparisons. — Medium sized (male 2.6 mm body length), ventral pro- jections on maxillary lobes of coxae II small; VAZQUEZ & COKENDOLPHER— NEW SPECIES OF HARVESTMAN 259 Table L — Appendage lengths (mm) of male holotype/male paratype of Guerrobunus vallensis new species. Segment Pedipalp Leg I Leg II Leg III Leg IV Trochanter 0.16/0.16 0.16/0.20 0.20/0.20 0.20/0.20 0.24/0.22 Femur 0.76/0.72 0.86/0.90 1.10/1.18 0.90/0.90 1.06/1.16 Patella 0.50/0.50 0.34/0.36 0.44/0.44 0.30/0.30 0.40/0.44 Tibia 0.48/0.46 0.54/0.60 0.94/1.00 0.64/0.64 0.94/0.98 Metatarsus — 0.60/0.68 0.90/0.98 0.84/0.84 1.16/1.24 Tarsus 0.50/0.50 0.54/0.58 1.04/1.16 0.60/0.62 0.74/0.80 Totals 2.40/2.34 3.04/3.32 4.62/4.96 3.48/3.50 4.54/4.84 coxae I with two tubercles anteriorly (larger in female), eyes present, corneas clear, retina unpigmented; ocular tubercle smooth; penis without paired apophysis on truncus dorsally. Guerrobunus vallensis new species appears to be closely related to Guerrobunus minutus but the former differs by the lack of low tubercles on the free tergites, the absence of pigment in the eyes and the total length of the body. The general structure of the penis of G. vallensis is similar to that of the male paratype of G. minutus. A detailed study of the paratype was not possible because the specimen could not be dissected, but the portion of the glans which is extending beyond the operculum ap- pears very similar to G. vallensis. The penis of Guerrobunus arganoi (Silhavy 1973, fig. 41) is also similar to that of G. vallensis. The differences between them are: the number of setae below the ventral plate, the stylus in G. vallensis is blunt with projections, it is pointed in G. arganoi with two lamella on the stylus; there are ten setae between blades in G. val- lensis whereas G. arganoi has eight. Other difference is: pedipalp of male G. vallensis has three seta-bearing tubercles on patella, in G. arganoi is one. Type data. — Male holotype, female allo- type and male paratype from Cueva del Dia- blo, Valle de Bravo, State of Mexico, Mexico; 25 April 1990, 1. Vazquez. Male holotype and female allotype deposited in the arachnologi- cal collection of Laboratorio de Acarologia of Instituto de Biologfa, UNAM. Male paratype deposited at the American Museum of Natural History. Description (measurements in mm). — Male: Total length (without chelicerae) 2.60, width 1.40; scute length (prosoma) 1.80, 1.40 wide at boundary with free tergites. Length of legs in Tables 1, 2. Anterolateral comers of cephalothorax each with a row of four obtuse tubercles, extending laterally (Fig. 2); thoracic tergites almost indistinct (Fig. 1), only visible in lateral view. Ocular tubercle rounded, not cone-shaped, close to anterior margin of pro- soma; eyes placed on each side of ocular tu- bercle (Fig. 1). Low hump behind ocular tubercle; free tergites without rows of small tubercles. Pedipalps (Figs. 4, 5) with spine- bearing ventrolateral tubercles: trochanter with one, femora with seven, patella with three, tibia with seven, tarsus with four. Max- illary lobes of second coxae not distinct, with one small tubercle on each, as in female (Fig. 9); coxae I with two tubercles anteriorly. Tar- sal segments 3:4: 5:5; distitarsus I with two segments, II with three segments (both males same). Pedipalp lengths in Table 1 . Color light red to orange, except leg tarsi and eyes white. Body and legs finely granulated. Penis (Fig. 6) with two visible parts: glans blunt, with lateral projections, tmncus cylindrical, oval in cross section, with a pair of sclerotized blades Table 2. — Leg lengths (mm) of the species of Guerrobunus. Taxa Leg I Leg II Leg III Leg IV Guerrobunus minutus (holotype female) 2.20 3.62 2.50 3.60 Guerrobunus vallensis (allotype female) 4.40 5.20 3.70 5.20 Guerrobunus vallensis (holotype male) 3.14 3.84 3.42 4.42 Guerrobunus arganoi (holotype male) 4.20 6.70 4.50 6.40 260 THE JOURNAL OF ARACHNOLOGY Figures 1-10. — Guerrobunus vallensis new species. 1-7, Male holotype. 1, Lateral view; 2, Dorsal view of prosoma (right comer), with detail of granulation; 3, Ventral view, genital operculum with penis; 4, Right pedipalp, lateral view; 5, Right pedipalp, medial view; 6, Distal part of penis, dorsal view; 7, Distal part of truncus with detail of setae, ventral view. 8-10, Female allotype. 8, Dorsal view; 9, Ventral view, genital operculum and coxae; 10, Distal end of ovipositor. VAZQUEZ & COKENDOLPHER— NEW SPECIES OF HARVESTMAN 261 Table 3. — Comparison of males of the species of Guerrobunus (scute length of G. arganoi obtained by measuring Silhavy 1973: fig. 42). Guerrobunus Guerrobunus Guerrobunus minutus vallensis arganoi Scute length 1.06 1.80 1.9 Total length 1.62 2.60 2.60 Pedipalp segment ratios 7:4:7:6 7:3:7:4 7:2:6:4 Pedipalp length 1.90 2.44 2.30 Tarsal segments 3:4:5:5 3:4:5:5 3:4:4:5 Distitarsus I:II 2:2 2:3 2:2 Eyes present/pigmented present/no color absent (— ventral plate); truncus with five pairs of tiny setae below the paired blades of ventral plate. Ventrally, between the blades, are ten long and thick setae in a triangular arrange- ment (Fig. 7). Penis 1.29 long: glans plus sty- lus 0.75 long, truncus 0.54 long; with four paired dorsal setae just below blades, and three pairs of ventral setae; six short spiny se- tae are on each side of truncus below glans. Stylus blunt, maximum width 0.20 (Fig. 7). Genital operculum 0.62 long, 0.48 wide, with 14 pairs of setae and one apophysis on each anterolateral comer (Fig. 5). Female: Total length 2.40; scute 1.54 long, 1.50 wide at the boundary with abdomen. Leg lengths as in Table 2. Anterolateral comers of prosoma each with a row of three obtuse tu- bercles (Fig. 3). General stmcture of prosoma and abdomen as in male. Spination of pedi- palps as in male (Figs. 4, 5). Tarsal segments: 3:4:5:5; distitarsus I and II with two segments each. Tubercles on coxae I more robust than in male (Fig. 9). Color light red, leg tarsi and eyes white. Genital operculum almost as wide as long with three or four spine-like apophy- ses on each anterolateral comer (Fig. 9). Dis- tal end of ovipositor (Fig. 10) with 29 long setae (each with 3-5 tips), arranged in four groups, three with 7 and one with 8 setae; two spine-like apophyses between setae groups. ACKNOWLEDGMENTS Dr. Norman L Platnick and Dr. Tila M. Pe- rez, curators of arachnology at the American Museum of Natural History, New York and the Laboratorio de Acarologia, Institute de Biologia, Universidad Nacional Autonoma de Mexico, Mexico, respectively, are thanked for the loan of the types of Guerrobunus minutus. LITERATURE CITED Goodnight, C.J. & M.L. Goodnight. 1945. Addi- tional Phalangida from Mexico. American Mus. Novit., 1281:1-17. Goodnight, C.J. & M.L. Goodnight. 1953. The op- ilionid fauna of Chiapas, Mexico, and adjacent areas (Arachnoidea, Opiliones). American Mus. Novit., 1610:1-81. Goodnight, C.J. & M.L. Goodnight. 1971. Opi- lionids (Phalangida) of the family Phalangodidae from Mexican caves. Assoc. Mexican Cave Stud. Bull., 4:33-45. Goodnight, C.J. & M.L. Goodnight. 1973. Opi- lionids (Phalangida) from Mexican caves. Assoc. Mexican Cave Stud. Bull., 5:83-96. Goodnight, C.J. & M.L. Goodnight. 1977. Lania- tores (Opiliones) of Yucatan Peninsula and Be- lize (British Honduras). Assoc. Mexican Cave Stud. Bull., 6:139-166. Goodnight, C.J. & M.L. Goodnight. 1983. Opili- ones of the family Phalangodidae found in Costa Rica. J. Arachnol., 11:201-242. Pickard-Cambridge, EO. 1904. Arachnida. Araneida and Opiliones. Biol. Central!- Americana, 2:1-560. Rambla, M. & C. Juberthie. 1994. Opiliones. Pp. 215-230, In Encyclopaedia Biospeologica. (C. Juberthie & V. Decu, eds.). Societe de Biospeo- logie, Moulis -Bucarest, Tome 1. Reddell, J.R. 1981. A review of the cavemicole fauna of Mexico, Guatemala, and Belize. Texas Mem. Mus., Bull., 27:1-327. Shear, W.A. 1977. The opilionid genus Neogovea Hinton, with a description of the first troglobitic cyphophthalmid from the Western Hemisphere (Opiliones, Cyphophthalmi). J. Arachnol., 3:165- 175. Silhavy, V. 1974. Cavemicolous opilionids from Mexico (Arachnida, Opiliones). Quad. Accad. Naz. Lined., Probl. Att. Sci. Cult. (1973), 171:175-194. Silhavy, V. 1977. Further cavemicolous opilionids fi*om Mexico (Arachnida, Opiliones). Quad. Accad. Naz. Lined., Probl. Att. Sci. Cult, 171:219-233. Manuscript received 22 July 1996, accepted 20 May 1997. 1997. The Journal of Arachnology 25:262-268 KAIRA IS A LIKELY SISTER GROUP TO METEPEIRA, AND ZYGIELLA IS AN ARANEID (ARANEAE, ARANEIDAE): EVIDENCE FROM MITOCHONDRIAL DNA William H. Piel and Karen J. Nutt: Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138 USA ABSTRACT. Various authors have offered three alternative hypotheses of phylogeny which suggest different sister groups to the orb-weaving spider genus Metepeira. In one case Kaira is sister genus to Metepeira, and Zygiella is sister to Kaira plus Metepeira’, in another case, Kaira is sister genus to Me- tepeira, but Zygiella is a tetragnathid, and thus unrelated at this level of analysis; and in the last case, Zygiella is close to Metepeira, but this time Kaira is not closely related. To resolve among these conflicting hypotheses, six species of orb- weaving spiders were sequenced for mitochondrial DNA encoding a portion of the 12S ribosomal subunit. These data were analyzed with data from two tetragnathid orb-weavers to estimate the phylogenetic relationships among these spiders and to determine which genus is a likely sister group to Metepeira. Phylogenetic analysis using parsimony supports the hypothesis that Kaira is a likely sister group to Metepeira and that Zygiella is in the family Araneidae rather than the family Tetragnathidae. Relationships among orb-weaving spiders are, in general, poorly understood (Codding- ton & Levi 1991). In particular, it is not known which genus within the araneids is most closely related to the genus Metepeira F.P. -Cambridge 1903. Such information is valuable to a phylogenetic analysis of Mete- peira (about 40) species because it uncovers ancestral character states and shows patterns of character evolution among species (Mad- dison et al. 1984). It is our intention in this paper to compare 12S mtDNA sequences of several selected taxa in order to determine which among them is the closest outgroup to Metepeira. Scharff & Coddington (in press) hypothe- size that Kaira O.P.-Cambridge 1889 and Me- tepeira are sister groups because both genera share the loss of the stipes and have a median apophysis with a pair of prongs and a toothed anterior margin (compare fig. 82 with fig. 127 in Levi 1977). Thus, we targeted Kaira as a potential sister group to Metepeira. Somewhat similar median apophyses are also found in Aculepeira Chamberlin & Ivie 1942 and Ama- zonepeira Levi 1989, but that of Kaira is the most similar. Genitalic and somatic characters in Amazonepeira and Aculepeira align them closer to Araneus Clerck 1757 rather than to Metepeira (Levi 1977, 1989, 1993). Simon (1895), who was one of the first ar- achnologists to discuss relationships among orb-weaving spiders in detail, did not consider Kaira and Metepeira to be closely related. His classification created four sub-families within what he called the Argiopidae (= Araneo- idea), including Argiopinae (= Araneidae), which contained 28 “groups”, two of which were Poltyeae and Araneae. The Poltyeae group contained Kaira’, the Araneae group consisted of four “series”, largely defined by eye arrangements. Many species which today are called Araneus, as well as some species affiliated to Larinioides Caporiacco 1934, were placed in series number 2. Metepeira and Zygiella F.O. Pickard-Cambridge 1902 were placed in series number 3. (Fig. 1, right column). Zygiella is another genus which we have targeted as a candidate sister group to Mete- peira. Scharff & Coddington (in press) agree with Simon that Zygiella is close to Metepeira based on their morphological cladistic analy- sis. Coddington (1990) suggests that Zygiella, which has a radix, distal hematodocha, and terminal apophysis, belongs to the Araneidae (Fig. 1, middle column) and not the Tetrag- nathidae. This placement is in keeping with three synapomorphies that are thought to unite the Tetragnathidae, yet are absent in Zygiella’. 262 PIEL & NUTT— /sC4/^, METEPEIRA AND ZYGIELLA MTDNA 263 (Levi) (Scharff & Coddington) (Simon) M I ^ 1 |m ^ >> K < — — > 1 ^ 1 y L II ^ II L Z X L K T T T 1 u ^ 'w 1 u 1 u Figure 1. — Schematic diagrams illustrating three hypotheses of relationships for six orb-weaving taxa; hierarchical relationships are depicted as nest- ed sets of Venn diagrams. Left column, hypothesis of Levi (1977, 1980); middle column, hypothesis of Scharff & Coddington (in press); right column, hy- pothesis of Simon (1895). Abbreviations: M, Me- tepeira\ K, Kaira; Z, Zygiella; L, Larinioides; T, Tetragnatha; U, Uloborus. apical tegular sclerites, loss of the median apophysis, and a conductor that wraps a free embolus (Hormiga et al. 1995). In contrast, Levi (1980) considers Zygiella and Metepeira not to be closely related. He placed the former in the Metine group of the Tetragnathidae based on the closely spaced eyes and the con- ical tibia (Fig. 1, left column). To help decide among the hypotheses of Levi (Fig. 1, left column), Coddington & Scharff (Fig. 1, middle column), and Simon (Fig. 1, right column), we sequenced 12S ri- bosomal mtDNA from two individuals repre- senting different species of Metepeira, and one individual from each of Kaira, Zygiella, Larinioides, and Uloborus Latreille 1809. These sequences were analyzed with Gillespie et al.’s (1994) data for Tetragnatha Latreille 1804 and Doryonychus Simon 1900 (family Tetragnathidae). Obviously, these eight taxa form an extremely limited sample, but the in- tention here is to help us select among the three main taxonomic hypotheses relating to Metepeira rather than to attempt a compre- hensive analysis of the Araneidae. METHODS The six female spiders chosen for mtDNA extraction, amplification, and sequencing were: Metepeira daytona Chamberlin & Ivie 1942, from Flagler Beach, Florida (29°37'N, 82°23'W); Metepeira minima Gertsch 1936, from Tamaulipas, Mexico (22°30'N, 99°4'W); Kaira alba (Hentz 1850), from Lake Lochloo- sa, Florida (29°37'N, 82°23'W); Zygiella atri- ca (C.L. Koch 1843), from Nahant, Massa- chusetts (42°25'38.7''N, 70°56'9. 1"W); Larinioides sclopetaria (Clerck 1757), from Cambridge, Massachusetts (42°20'N, 7r6'W); and Uloborus glomosus Walckenaer 1842, from Sherman, Connecticut (41°34'30"N, 73°31'16"W). Specimens were collected in 80% or 100% ethanol. Vouchers were depos- ited at the Museum of Comparative Zoology. Tissue for extraction was dissected primar- ily from the prosoma: the carapace was lifted away, tissues were removed, and in many cases the carapace replaced so that the speci- men appeared unaltered. For some smaller specimens, muscle fibers were also taken from the chelicerae and femora. Care was taken to exclude the cuticle which, if present, could hinder amplification (J.K. Wetterer pers. comm.). Using chilled glass homogenizers, tissues were ground twice in 100 p.1 of 50 mM Tris- Cl, 20 mM EDTA, and 2% SDS. To digest the proteins, the extractions were incubated with 2 p.1 of 100 ng/ml proteinase K in a 60 °C oven for 1 h. To remove cell walls and resid- ual ionic compounds, 100 |xl of saturated NaCl were added. The extractions were cooled on ice for 30-70 min and then centri- fuged for 15 min at 4 °C. The supernatant was retained and the DNA was precipitated with 100% EtOH, washed with 70% EtOH, dried in a centrifuge under vacuum, and resuspend- ed in 100 p.1 IxTE (lOmM Tris-HCl and 1 mM EDTA). A 257 bp region of the third domain of the 12S ribosomal subunit was amplified and se- quenced for most taxa using primers 12St-L and 12Sbi-H (Croom et al. 1991). Mitochon- drial DNA from U. glomosus failed to work with 12St"L, so 12S-U [a degenerate arthro- pod-specific primer designed by D. Fitzpatrick (5 ^ -TGTTT( AT)( AGT)TA ATCG A( ATC)( AT) (ACT)T(AC)CACG-3')] was used instead. Two p.1 of template were used in 100 p.1 PCR reactions (50 mM KCl; 10 mM Tris-HCl; 0.1% Triton® X-100; 2.5 mM MgC^; 0.5 p.M of each primer; 2.5 units of Taq; and 0.2 mM dNTP) and cycled 30-35 times (45 sec at 94 264 THE JOURNAL OF ARACHNOLOGY Table 1. — Genetic distances among different species of orb- weavers. For each pairwise comparison, corrected percent distances [based on the Kimura two-parameter model (Li et al. 1985) and generated by Heap Big (Palumbi, unpub. program)] appear above the diagonal, percent transversions below the diagonal. Column headings, M. day = Metepeira daytona, M. min = Metepeira minima, K. alb = Kaira alba, Z. atr = Zygiella atrica, L. scl = Larinioides sclopetaria, U. glo = Uloborus glomosus, D. rap = Doryon- ychus raptor, T. per = Tetragnatha perreira. M. day M. min K. alb Z. atr L. scl U. glo D. rap T. per M. daytona — 13 18 31 30 45 49 44 M. minima 3 — 21 35 28 45 57 48 K. alba 9 7 — 27 25 44 53 41 Z. atrica 18 18 18 — 25 39 49 32 L. sclopetaria 17 14 16 17 — 44 39 36 U. glomosus 24 23 27 24 27 — 43 42 D. raptor 27 27 28 26 25 24 — 28 T. perreira 23 23 23 18 22 26 16 — °C; 60 sec at 42 °C; 90 sec at 72 °C). The PCR products were purified on a low melt agarose gel: bands corresponding to DNA of the appropriate length were cut from the gel, and DNA was isolated from the agarose using phenol or spin columns (QIAquick, by QIA- GEN®). The PCR product was sequenced in both directions using DyeDeoxy® termination (Perkin-Elmer Kit) with the same primers used in amplification. Sequence products were purified with CENTRI-SEP columns (Prince- ton Separations, Inc.) and then run on a ABI 370A autosequencer (Applied Biosystems, Inc.). Chromatogram sequence data generated by the autosequencer were edited by eye using SeqED (Applied Biosystems, Inc.). Sequence data collected by Gillespie et al. (1994) using the same primers on two Ha- waiian tetragnathid species, Tetragnatha per- reira Gillespie 1991 from Oahu and Doryon- ychus raptor Simon 1900 from Kauai, were added to our data set and aligned using Clustal V (Higgins et al. 1992). The resulting align- ment was further adjusted by hand and cropped to form a character matrix using SeqApp (Gilbert 1994). Corrected pairwise percent distances based on the Kimura two- parameter model (Li et al. 1985) were calcu- lated using the program Heap Big (Palumbi unpub. program). An exhaustive search for the most parsimonious tree and bootstrap analysis were performed using PAUP (Swofford 1991) on sequence characters for all eight species, holding U. glomosus as the outgroup. Trees were compared and manipulated with Mac- Clade (Maddison & Maddison 1992). RESULTS The resulting character matrix is 208 bases long (Fig. 2). An exhaustive search using PAUP (Swofford 1991) yields two most par- simonious trees, each 227 steps long, C.I. with all characters, 0.75; C.I. with uninformative characters excluded, 0.67; and R.L of 0.54. The two trees disagree only in whether L. sclopetaria is more closely related to K. alba plus Metepeira or whether it is more closely related to Z. atrica. The strict consensus of these two trees is illustrated in Fig. 4. Al- though pairwise genetic distances (Table 1) are quite high, skewness test statistic (gi) cal- culated by PAUP is —0.81, which is statisti- cally significant {P < 0.01), indicating that there is, nonetheless, strong phylogenetic sig- nal (Hillis & Huelsenbeck 1992). Further- more, an exhaustive search with U. glomosus excluded results in a single most parsimonious tree with Z. atrica most closely related to Me- tepeira plus K. alba — a result that is still com- patible with Fig. 4. Of the 76 unambiguous changes on the tree (i.e., character state changes that optimize to a single, specific branch segment), 3 1 are tran- sitions (purine to purine or pyrimidine to py- rimidine) and 45 are transversions (purine to pyrimidine or pyrimidine to purine). This par- adoxically low ratio of transition to transver- sion events increases with decreasing branch lengths (Fig. 3) and therefore is evidence for multiple hits and saturation between distant relatives (Simon et al. 1994). However, a tran- sition to transversion ratio of 0.69 is still with- PIEL & NUTT— METEPEIRA AND ZYGIELLA MTDNA 265 Metepeira daytona Metepeira minima Kaira alba Zygiella atrica Larinioides sclopetaria Uloborus glomosus Doryonychus raptor Tetragnatha perreira .10 .20 .30 .40 TACTCTTATTTAAA--TCTTATATACCTCCATCTTAAGAATTAATATCTA ... .T. C. . .G C. . .T. . A.T. T. . .G. .A-A. C. . .TA. .AC. ......... -AT ... .T. .............. A-AT C. . .T.A.A. . . . .AGT. ....... .G AA. .- T-.C.TC. . . . .TGG. ... .T.-.T. ..... .T. . . .G.A-.C- AT. .C.-A. C. . .T. . .A. . .T.-.T G. .G. .-.T-. . . .AG. .T.-A. .50 .60 .70 TATTCTCTTCTAAACAGAAATTC - .80 .90 .100 .110 .120 -TAAAAAGTTAGGTAAAGGTGTAGACTACATAAGAGTTTATGTGGGTTACAATAAA . .CC.CT. , .T. . . , .G. ..... . . . .A. .C.GGG. . C. .A, .A C. , .A.T. . .C. C.T. . . . .A. . . . . .G. . A.A.-.T.-T.C. . .A.T. . . .AT-T. . . .T. . . . TT, . .TA. .A. .TGA. .T. . . . . .T ATA. T . T . -T . CTTA . T . .T. .T-T. , , .T.T. . . . . A.T. , . . .A. .A.A.GA. . .A. . . . .T. . . . A. . .AC-AAT.C. , .A.T. .TA.T-A. . TA. . . . . .C. . . . . . . .ATT. .A. , .TTA. .AGAGGA. .T. . . . . .T ATA. A.-. AT. C. .CA.T. .TA.T-A. . .T, . . . .CCA. . . .A.AA. .TA. . . . .T. . . . A. A.T.-. AT. C. . .A. A. . .A.T-C. . T A TT, . . .A. .ACA.AA. .TA. . . . .T .130 .140 .150 .160 .170 .180 .190 .200 TTCTATTTAAGAATATATAATTAAAATTTTAT-TTGAAAAAGGATTTGTAAGTAAAT-TAAAAATAATATTCTTTATTG .CT. .T. .. .A- .A. ..... .T.-. . .G. ........ .C. . .A . .AG TA A. .T.-C G. ..... .A .T.- A.T A . . .G.A. .TA.T.A A. .A. . -AA G. . . . . . .AACT .-.T. . .C. .T T.AA. . ..AG..T.T.T TA.AA..- .A.T. ....T.T. TTT .... T ... . TAA . . A . . .G. . .AT. . .A. TTT. A. . . .AT, .T. AAT.T -AT. . .T-. .A. A.T. A. . . . . .GC.TA. .GTA, . . .TCGCAT.-A.A. . . .C. . . . . .A. . . . . .T.A-.TTTTT. . .A.AAAAAGT. . . .G. .CG.GTTAA.T.TAA. .TC-AA .A. ... . .TTA-.TT. .T .AAAAT. . .A. . . .AG. . .T. . A.TA. AA.A. .AAA. Figure 2. — Matrix of 208 characters from aligned 12S ribosomal mtDNA sequences. Data for eight orb- weaving taxa are represented, two of which {Doryonychus raptor and Tetragnatha perreira) were pub- lished in Gillespie et al. (1994). in the range of other comparable and success- ful phylogenetic analyses, such as 0.61 for the analysis of tetragnathid relationships by Gil- lespie et al. (1994). DISCUSSION Our data support the Scharff and Codding- ton hypothesis (compare Fig. 4 with Fig. 1, middle column) and thus provide evidence that Kaira is, indeed, a likely sister-group to Metepeira. Despite the fact that Metepeira is a morphologically homogeneous taxon with a restricted distribution, thus presumably with a relatively recent inception, the within-Mete- peira distances are not much shorter than those between Metepeira and Kaira (Table 1), Furthermore, the Kaira-Metepeira clade is supported by 94% of 1000 bootstrap replicates and eight unambiguous apomorphies (Fig. 4). Nonetheless, the branch lengths between clades seem more evenly spaced than what one might at first expect, given that some unite closely related taxa, whereas others unite distantly related taxa. However, this may merely reflect multiple substitutions, in which long genetic distances are vastly underesti- mated when new mutations occur at the same sites as the old mutations. Evidence for this occurrence can be seen in the attenuation of A B C D 1 {K. alba, (M. daytona, M. minima)) (L. sclopetaria, Z. atrica. A) ({T. perreira, D. raptor), B) ([/. glomosus, C) 0 0.5 1 1.5 2 2.5 Ratio of transitions to transversions. Figure 3. — Ratio of unambiguous transitions to unambiguous transversions for increasingly inclu- sive clades as calculated by MacClade (Maddison & Maddison 1992). Clade A includes Kaira alba, Metepeira daytona, and M. minima; clade B in- cludes Larinioides sclopetaria, Zygiella atrica, and clade A; clade C includes Tetragnatha perreira, Doryonychus raptor, and clade B; clade D includes Uloborus glomosus and clade C. 266 THE JOURNAL OF ARACHNOLOGY CO S Co 0 s 1 Co g 9k s I Co S CJ S' S' .2 Q ?3 I 18.5 93% I 1 CJ CO Co :i O SS ?3 S :§ 11.5 14.5 (4) « 2 « s S t .5 t I 2 t I 11.5 L*(9^ 76%^ij M~97^ 12.5 7.5 16 Tetragnathidae 76% 8.5 Araneidae Uloboridae I Figure 4, — Strict consensus tree of the two most parsimonious phylogenetic trees from 208 bases of the 12S ribosomal mtDNA subunit (tree length = 223+ steps; C.I. using all characters = 0.79; C.I. using informative characters only = 0.75). Figures adjacent to each branch indicate the number of unambiguous character changes averaged between the two most parsimonious trees. The figure in parentheses indicates the number of unambiguous character changes between where L. sclopetaria or Z. atrica branch from the main stem in either shortest tree. Percentages are bootstrap values for each node from 1000 replicates. the transition to transversion ratio when mea- sured over increasingly inclusive clades (Fig. 3). Although transitions occur more frequently than transversions, accumulation of transver- sions in older, longer branches will mask the activity of transitions (Simon et al. 1994). The pronounced attenuation of the transition to transversion ratio in our data suggest that the Kaira is actually closer to Metepeira, yet far- ther from the other taxa, than what the tree would appear to show. The same can be said for the separation between the araneids and the tetragnathids. The close relationship between Kaira and Metepeira, as evidenced from our results, in- dicates that the shared flagellated median apophysis, as well as other genitalic charac- ters, are likely to be homologous structures. Despite this particular similarity, Kaira and Metepeira share few other morphological fea- tures. Kaira has evolved numerous autapo- morphies as a result of its highly specialized PIEL & NUTT— METEPEIRA AND ZYGIELLA MTDNA 267 predatory behavior. Convergent with Masto- phora Holmberg 1876, Kaira has forgone orb- weaving, and is thought to emit pheromones that mimic those of female moths (Levi 1994; Stowe 1986). Thick, stubby setae on Kaira' s legs are presumably used to grab moths in flight, while a large array of tubercles on Kaira' abdomen are thought to conceal or protect the exposed spider while it is in its hunting posture. In contrast, Metepeira has neither specialized leg setae nor abdominal tu- bercles, and it weaves a very distinctive web which combines orb and scaffolding with as- sociated aerial retreat. Identifying the sister group to Metepeira can help clarify phylogenetic structure and character evolution within the genus. Levi (1977) divided Metepeira species north of Mexico into two groups: M. labyrinthea and M. foxi. Species in the former group have a white line on a black sternum and a short keel on the median apophysis. Species in the latter group have a uniform sternum and a distal tu- berculate keel on the median apophysis. Levi (1977) admitted that it “is difficult at present to decide which of these species groups is the derived and which the more primitive”. In- deed, one needs an outgroup in order to de- termine which species group contains species arising basally and retaining symplesiomorph- ic characters, and which species group con- tains species arising more distally and sharing synapomorphic characters. With Kaira as an outgroup to Metepeira, we can infer that the character states that de- fine the M. foxi species group are primitive, and thus species in this group may arise ba- sally within the genus Metepeira. Indeed, the distal tuberculate keel on the median apoph- ysis is similar to modifications in the median apophysis of Kaira (compare figs. 82, 91-127 in Levi 1977). Furthermore, Kaira lacks the white line on a black sternum as seen in the M. labyrinthea species group. Also, within the M. foxi species group, M. daytona is probably the most basal species because the ratios be- tween patella-tibia and metatarsus-tarsus arti- cles are the same as they are in Kaira (about 1.1:1); whereas in all other known Metepeira species the ratio is about 0.9:1. Thus, with Kaira as the outgroup, our results support the relatively basal origins of species in the M. foxi species group. However, since this group is defined by exclusively symplesiomorphic characters, we cannot infer that it is mono- phyletic. Nine unambiguous synapomorphies support the inclusion of Z. atrica within the Araneidae (Fig. 4). Forcing Z. atrica into the Tetragnath- idae costs four additional steps. In addition, 1000 bootstrap replicates using PAUP support the araneid clade 76% of the time. Thus, our data disagree with Levi (1980) and others who believe that Zygiella is a tetragnathid. However, we should mention that the monophyly of Zygiella is uncertain. On the one hand, the vacant sector in the orb web, the compact eye region, and the dorsoventral- ly flattened oval abdomen with its character- istic markings, seem to unite Zygiella species (Levi 1974). On the other hand, the inconsis- tency in the presence of a scape, terminal apophysis, paracymbium shape, tooth on the palpal endite, and seta on the palpal patella, put monophyly of the genus into question (Levy 1986). Levi (1974) argues that the re- markable diversity in Zygiella genitalia fails to break apart the genus because many incon- sistent characters overlap one another. For ex- ample many species that lack a scape still share a derived ventral apophysis of the te- gulum with other species that have a scape (Levi 1980). Thus, while it is still possible that Zygiella is paraphyletic, and while it is possible that some Zygiella species are, in fact, tetragnathids, our data argue that at least the type species for the genus, Z. atrica, ap- pears to be an araneid. ACKNOWLEDGMENTS A million thanks go to N.E. Pierce for use of her equipment and laboratory facilities. Thanks also to D.A. Fitzpatrick, H.W Levi, and E.O. Wilson for their advice and support; to M. Stowe for supplying spider tissue; to D. Campbell, J. Coddington, G. Hormiga, and D. McHugh for commenting on the manuscript; and to R.G. Gillespie for sending us her data matrix. LITERATURE CITED Coddington, J.A. 1990. Ontogeny and homology in the male palpus of orb-weaving spiders and their relatives, with comments on phylogeny (Ar- aneoclada: Araneoidea, Deinopoidea). Smithson- ian Contr. ZooL, 496: 1-52. Coddington, J.A. & H.W. Levi. 1991. Systematics and evolution of spiders (Araneae). Annu. Rev. Ecol. Syst., 22:565-592. 268 THE JOURNAL OF ARACHNOLOGY Groom, H.B., R.G. Gillespie & S.R. Palumbi. 1991. Mitochondrial DNA sequences coding for a por- tion of the RNA of the small ribosomal subunits of Tetragnatha mandibulata and Tetragnatha ha- waiensis (Araneae, Tetragnathidae). J. Arachnol., 19:210-214. Gilbert, D.G. 1994. SeqApp: Mac sequence editor, version 1.9a. Indiana University, Bloomington, Indiana. Gillespie, R.G., H.B. Groom & S.R. Palumbi. 1994. Multiple origins of a spider radiation in Hawaii. PNAS USA, 91:2290-2294. Higgins, D.G., A.J. Bleasby & R. Fuchs. 1992. GlustalV: Improved software for multiple se- quence alignment. GABIOS, 8:189-191. Hillis, D.M. & J.P. Huelsenbeck. 1992. Signal, noise, and reliability in molecular phylogenetic analyses. J. Hered., 83:189-195. Hormiga, G., W.G. Eberhard & J.A. Goddington. 1995. Web-construction behaviour in Australian Phono gnatha and the phylogeny of Nephiline and tetragnathid spiders (Araneae: Tetragnathi- dae). Australian J. Zool., 43:313-364. Levi, H.W. 1974. The orb- weaver genus Zygiella (Araneae: Araneidae). Bull. Mus. Gomp. Zool., 146:267-290. Levi, H.W. 1977. The orb-weaver genera Metepei- ra, Kaira and Aculepeira in America North of Mexico (Araneae: Araneidae). Bull. Mus. Gomp. Zool., 148:185-238. Levi, H.W. 1980. The orb- weaver genus Mecyno- gea, the subfamily Metinae and the genera Pach- y gnatha, Gleno gnatha and Azilia of the subfam- ily Tetragnathinae north of Mexico (Araneae: Araneidae). Bull. Mus. Gomp. Zool., 149:1-74. Levi, H.W. 1989. The neotropical orb- weaver gen- era Epeiroides, Bertrana and Amazonepeira (Araneae: Araneidae). Psyche, 96:75-99. Levi, H.W. 1993. The orb-weaver genus Kaira (Araneae: Araneidae). J. Arachnol., 21:209-225. Levy, G. 1987. Spiders of the genera Araniella, Zygiella, Zilla, and Mangora (Araneae, Aranei- dae) from Israel, with notes on Metellina species from Lebanon. Zool. Scr., 16:243-257. Maddison, W.P., M.J. Donoghue & D.R. Maddison. 1984. Outgroup analysis and parsimony. Syst. Zool., 33:83-103. Maddison, WP. & D.R. Maddison. 1992. Mac- Glade: Analysis of phylogeny and character evo- lution, version 3.0. Sinauer, Sunderland, Massa- chusetts. Roewer, G.F 1942. Katalog der Araneae von 1758 bis 1940. Bremen, 1:1-1040. Scharff, N. & J.A. Goddington. In press. A phy- logenetic analysis of the orb- weaving spider fam- ily Araneidae (Arachnida, Araneae). Zool. J. Linn. Soc., 00:00-00. Simon, G., F. Frati, A. Beckenbach, B. Grespi, H. Liu, & P. Flook. 1994. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved poly- merase chain reaction primers. Ann. Entomol. Soc. America, 87:651-701. Simon, E. 1895. Histoire naturelle des Araignees. 2nd ed., Paris. Stowe, M. 1986. Prey specialization in the Ara- neidae. Pp. 101-131, In Spiders: Webs, Behavior and Evolution. (WA. Shear, ed.). Stanford Univ. Press, Stanford, Galifomia. Swofford, D.L. 1991. PAUP: Phylogenetic analysis using parsimony. Illinois Nat. Hist. Surv., Gham- paign, Illinois, version 3.0s. Manuscript received 26 July 1996, accepted 22 April 1997. 1997= The Journal of Arachnology 25:269-287 SCHARFFIA, A REMARKABLE NEW GENUS OF SPIDERS FROM EAST AFRICA (ARANEAE, CYATHOLIPIDAE) Charles E. Griswold: Department of Entomology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118 USA ABSTRACT. The new genus Scharffia (Araneae, Cyatholipidae), comprising the new species Scharffia chinja, Scharffia holmi, Scharjfia nyasa and Scharjfia rossi, is described. Discovered in southern Africa near the end of the last century (Simon 1894; Cambridge 1903), the Cyatholipidae comprise rich faunas in the cool-temperate southern latitudes of Af- rica (Griswold 1987) and Australasia (Forster 1988). They are typical denizens of the “Af- romontane” forests (White 1978; Griswold 1991) of the mountains and Cape coasts of South Africa and, as is the case with many other animals and plants, their occurrence in the moist, montane forests making up the “Afromontane archipelago” in tropical Africa should come as no surprise. Cyatholipids have recently been described from Madagascar (Griswold 1997): herein I describe the first cy- atholipids recorded from tropical Africa. Most collection records suggest that Scharffia favor wet forests. They are common in montane forests (i.e., above 800 m elevation) and typi- cally absent from nearby lowland forests (though at least S. chinja new species has been collected beneath 3(X) m elevation). Scharffia rossi new species was collected in dry savanna far from forest, and, like Cyatholipus hirsutis- simus Simon 1894 and Ulwembua denticulata Griswold 1987 (Griswold 1987), indicates that the family is not entirely restricted to forests. As is typical of cyathohpids, Scharffia hang beneath sheet webs (Figs. 2-4; Davies 1978; Forster 1988; Griswold et al. in press) and were rarely collected away from webs (e.g., in pitfalls or by sifting). The function of the elongate, an- nulate abdominal petiole (Fig. 1) is unknown; but, to the casual observer, it renders the spiders remarkably similar to ants. The awl-shaped ab- domen of the S. chinja population at Mazumbai in the West Usambara Mountains of Tanzania (Fig. 19) makes them strikingly similar to Cre- matogaster ants. Nevertheless, this resemblance is not enhanced by hanging beneath sheet webs, nor do spiders collected on beating sheets move Uke ants: mimicry is a doubtful explanation for their remarkable abdominal modification. The sclerotized petiole may function in carapace-ab- domen stridulation, as recorded in the cyatho- hpid sister group Synotaxidae (Forster, Platnick & Coddington 1990; Griswold et al. in press). METHODS Prior to examination with a Hitachi S-520 Scanning Electron Microscope all structures were critical point dried. Vulvae were cleaned by exposure to trypsin, bleached in 5% sodi- um hypoclorite (Chlorox®), stained with Chlorazol Black, and mounted in Hoyer’s Me- dium for examination and photography. Ex- amination was via Wild M5Apo and Leitz Or- tholux II microscopes; and photography of vulvae was by an Olympus PM- 1 OAK at- tached to the Leitz Ortholux II. Small struc- tures were examined in temporary mounts as described in Coddington (1983). Abbreviations are fisted in Table 1. All mea- surements are in mm. For the key and diagnoses the ratio of the length of the palpal bulb (LPB)/ length of the median lobe of the tegulum (MLT) is based on the measurements: LPB = distance from distal margin of the apical lobe (A) of the tegulum to the proximal-most extent of the em- bolic curve; MLT = distance from distal margin of the apical lobe (A) of the tegulum to the proximal margin of the median lobe. Specimens measured were chosen to encompass largest and smallest individuals. 269 270 THE JOURNAL OF ARACHNOLOGY Figure 1. — Scharffia rossi new species, holotype male, lateral view. TAXONOMY Cyatholipidae Simon 1894 Cyatholipeae Simon 1894: 711, based on Cyatholipus hirsutissimus Simon 1894. Roewer 1942: 967. Cyatholipinae, Wunderlich 1978: 33. Teemenaaridae Davies 1978: 42, based on Teeme- naarus silvestris Davies 1978. Cyatholipidae Platnick 1979: 116. Brignoli 1983: 231. Griswold 1987: 501. Forster 1988: 7. Plat- nick 1989: 181. Platnick 1993: 172. Wunderlich 1993: 234. Diagnosis. — Colulate, entelegyne araneoids that share with the Synotaxidae a cup- shaped paracymbium (Figs. 27, 35) and posteriorly broadly truncate sternum, and differing in hav- ing a retromedian cymbial process (Figs. 12, 27) and very broad posterior respiratory groove (Figs. 10, 21). For full description see Griswold (1987) and Forster (1988). Scharffia new genus Type species. — Scharffia chinja new species. Etymology. — Named in honor of Nikolaj Scharff, Afromontane arachnologist and col- lector of many new and interesting Cyatholi- pidae; gender feminine. Note. — Scharffia has been previously men- tioned as “an undescribed genus occurring in montane forests from Malawi to Kenya” related to the Malagasy Alaranea (Griswold 1997, p. 82). Diagnosis. — Distinguished from all Cy- atholipidae by having the sternum elongate, prolonged between coxae IV, with length greater than 1.15X width (Figs. 8, 21, 36), and from all genera except Alaranea by having the anterior portion of abdomen of both sexes forming a sclerotized, annulate petiole, in most species elongate (Figs. 11, 16-22). Description. — Total length 2.25-3.25. Car- apace typically trapezoidal or diamond- shaped in dorsal view (Fig. 20), may be prolonged posteriorly (Fig. 32), length 1. 58-2.43 X width, posterior margin truncate, low, maximum height 0.35-0.57 X width, texture rugose (Fig. 9); thoracic fovea typically shallow, diamond- shaped to indistinct; ocular area with PER width 2.18-2.93X OAL, 2.25-2.80X OQP, OQP 0.87-1. 20X OQA; diameter AM 1.09- 1.87X PM, distance PM-PL 1.20-2.25X PM diameter; clypeal height 1.86-2. 80 X AM di- ameter, cheliceral length 1.35-2.54X clypeal height; chelicerae unmodified or with small basal protuberance, promargin with four, retro- margin with three teeth (Fig. 6). Sternum ru- gose to pustulate (Fig. 8), length 1. 15-1. 58 X width, coxae surrounded by pleural and sternal sclerotizations (Figs. 1, 5, 8). Abdomen oval to triangular, with short, slender setae, bases of anterior setae unmodified, sclerotized from ep- igastric furrow to and surrounding pedicel to form short-to-long annulate petiole (Figs. 11, GRISWOLD— 5C//ARFF/A 271 Figures 2-4. — ^Webs of Scharffm chinja new species, from Amani. 2, Webs on tree buttress (Scale bar = 10.0 cm); 3, Web, close up (Scale bar = 5.0 cm); 4, Underside of web with spider (arrow) (Scale bar =1.0 cm). 26), spinnerets surrounded by yellow-brown sclerotization with dark radial streaks (Figs. 21, 36). Legs unmodified, long (Figs. 1, 18) to ex- tremely long (Fig. 43), ratio 1 -2-4-3, female femur I length 2.42-4.67 X carapace width, male 2.51-9.48. Male palpus with retrolateral cymbial process (RMP) pointing ventrad (Figs. 12, 27), smaller than paracymbium (PC); pal- pal bulb (Figs. 14, 27-29) with dentate median lobe (MLT), apex (A) a small, smooth to pus- tulate lobe; conductor (C) median, longitudinal, simple (Figs. 28, 29, 57) or with accessory pro- cess (Figs. 14, 52), smooth; embolus (E) thick, making simple curve, origin apical between 10-11 o’clock, ridged; parembolic process (PP) present (Figs. 14, 15, 53) or absent (Figs. 28, 56), thick and fleshy with a median attenuate projection, lacking teeth, with or without pus- tules; sperm duct with curlicue near embolic base. Epigynum (Figs. 23-26) with scape (S) and median hood (MH) with slender septum between copulatry openings (CO), atrial fur- rows (AT) extending behind scape. Vulva (Figs. 37-40) with sclerotized, simple, narrow to hemispherical lateral afferent duct (AD), fer- tilization duct (FD) posterior to spermathecal head (HS). Composition.- — Four species. Distribution. — East Africa from Malawi to Kenya (Fig. 58). KEY TO SPECIES OF SCHARFFIA 1 Abdomen with petiole length greater than 0.24 of carapace length (Figs. 1, 18, 20) 2 - Abdomen with petiole length less than 0.17 of carapace length (Figs. 41-43) .......... .nyasa 2(1) Posterior portion of carapace elongate, forming parallel-sided neck, carapace length greater than twice width (Figs. 1, 32); embolus without parembolic process (Figs. 30, 34); conductor simple; epigynal scape twice as long as wide (Fig. 33) 3 - Carapace diamond-shaped in dorsal view (Figs. 19, 20), posterior portion tapering, carapace length less than twice width; embolus with parembolic process (Fig. 44); conductor double; epigynal scape much wider than long (Fig. 46) . . chinja 3(2) Length palpal bulb less than 2X that of the median lobe of the tegulum (MLT), tegulum nearly hidden between MLT and embolus (Figs. 30, 56). rossi - Length palpal bulb greater than 2.5 X MLT, tegulum clearly visible between MLT and embolus (Figs. 28, 34) . ................. ..... holmi 272 THE JOURNAL OF ARACHNOLOGY Table 1. — List of anatomical abbreviations used in the text and figures. A apical lobe of tegulum AD vulval afferent duct AER anterior eye row AL anterior lateral eyes AM anterior median eyes AT epigynal atrium C conductor CB cymbium CO copulatory opening E embolus EF epigastric furrow FD fertilization duct HS spermathecal head LPB length palpal bulb MH epigynal median hood ML epigynal median lobe MLT median lobe of tegulum MS epigynal median septum OAL ocular area length OQA ocular quadrangle, anterior OQP ocular quadrangle, poseterior PC paracymbium PER posterior eye row PL posterior lateral eyes PM posterior median eyes PP parembolic process RMP retromedian cymbial process S epigynal scape ST subtegulum T tegulum TL ventromedian tegular lobe Scharffia chinja new species (Figs. 2-23, 25, 38, 40, 44-46, 58) Types.— Male holotype and female para- type from intermediate rain forest at Uzungwa Scarp Forest Reserve above Chita village, elev. 1050 m, Uzungwa Mts., Iringa Region, Tanzania, 5 November 1984 (N. Scharff) (ZMUC). Etymology. — The specific epithet is an ar- bitrary combination of letters. Diagnosis. — Distinguished from nyasa new species by having the abdominal petiole great- er than 0.24 carapace length (Figs. 18, 20); males distinguished from rossi new species and holmi new species by having a parembolic process and double conductor (Figs. 44, 45); females distinguished from holmi by having a broad scape (Fig. 46) and hemispherical af- ferent ducts (Figs. 38, 40). Description. — Male (holotype): Total length 2.64. Carapace, clypeus, chelicerae, sternum, labium, and palpal coxae dark red- brown, unmarked except for dusky macula- tions on clypeus; palpi dark yellow-brown, unmarked; coxae, trochanters, and legs yel- low-brown, unmarked except for subbasal brown annulus on femur IV; abdomen dark gray, dorsum with narrow longitudinal and broad transverse white markings forming cross. Carapace 1.21 long, 0.61 wide, 0.29 high, prolonged posteriorly to meet abdominal petiole; PER 0.38 wide, AER 0.37 wide, OAL 0.17; ratio AM:AL:PM:PL, 1.6:L2:1.0:1.2, PM diameter 0.05. Clypeus 0.18 high, chelic- erae 0.26 long. Sternum 0.58 long, 0.47 wide; labium 0.10 long, 0.16 wide; palpal coxae 0.18 long, 0.14 wide. Leg measurements (fe- mur + patella + tibia + metatarsus + tarsus = [Total]): I: 2.64 + 0.25 + 2.23 + 2.13 + 0.91 = [8.13]; II: 1.81 + 0.23 + 1.57 + 1.49 + 0.72 = [5.82]; III: 0.87 + 0.17 + 0.64 + 0.62 + 0.40 = [2.70]; IV: 1.32 + 0.19 + 1.06 + 0.87 + 0.42 = [3.80]; Palp: 0.26 + 0.10 + 0.08 + (absent) + 0.26 = [0.70]. Palp (Figs. 12-15, 44, 45) with RMP narrowly triangular, PC narrow, deeply concave in lateral view; tegulum apex pustulate, MLT large, convex, dentation restricted to narrow longitudinal band; C large, with small narrow secondary process; PP present, lacking pustules. Variation: (n = 7). Total length 2.34-2.89; ratios of carapace length/width 1.74-2.00, height/width 0.35-0.52, PER/OQP 2.37-2.64, PER/OAL 2.19-2.80, OQP/OQA 0.87-1.07, diameter AM 1.18-1.60 times PM; ratios of clypeal height/ AM diameter 2.12-2.61, chel- iceral length/clypeal height 1.35-1.87; ratio of sternum length/width 1.15-1.46; ratio of fe- mur I length/carapace width 4.00-5.01. The shape of the soft part of the abdomen ranges from nearly round (Figs. 17, 22) to triangular (Figs. 16, 18, 20) to heart- to awl-shaped (Fig. 19: dorsal view of Mazumbai specimen). Markings also vary greatly: the dorsum may be all dark, have lateral light spots (Fig. 17) or a narrow to broad transverse median band (Fig. 22); a narrow to broad longitudinal me- dian band may be present anteriorly (Fig. 19), separate from transverse band (Fig. 20) or connected to it to form a light cross (Fig. 16). Female (paratype): Total length 2.58. Markings as in male except white markings of abdomen not forming cross, longitudinal dor- sal mark attenuate anteriorly, with anterolat- GmSWOLD—SCHARFFIA 273 Figures 5-11. — Scharffia chinja new species, female, from Uzungwa. 5, Carapace, lateral; 6, Mouth- parts, ventral; 7, Face; 8, Sternum and petiole, ventral; 9, Carapace, dorsal; 10, Spinnerets and posterior spiracle (arrows); 11, Abdominal petiole, lateral. (Scale bars for Figs. 5-8, 11 = 100 jxm; Fig. 9, 250 jam; and Fig. 10, 50 |xm.) eral faint white spot and median lateral trans- verse band. Structure as in male; carapace 1.17 long, 0.58 wide, 0.28 high; PER 0.39 wide, AER 0.38 wide, OAL 0.17; ratio AM: AL:PM:PL, 1.6: 1.2: 1.0: 1.4, PM diameter 0.05. Clypeus 0.17 high, chelicerae 0.33 long. Sternum 0.67 long, 0.44 wide; labium 0.11 long, 0.14 wide; palpal coxae 0.20 long, 0.16 wide. Leg measurements (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 1.72 274 THE JOURNAL OF ARACHNOLOGY Figures 12-15. — Scharffia chinja new species, from Amani, right male palpus. 12, Retrolateral; 13, Prolateral; 14, Ventral; 15, Parembolic process. A = apical lobe of tegulum, C = conductor, CB = cymbium, E = embolus, PP = parembolic process, RMP = retromedian cymbial process, ST = subte- gulum, T = tegulum, TL = ventromedian tegular lobe. (Scale bars for Figs. 12-14 = 60 fxm. Fig. 15 = 15 |jLm.) + 0.23 + 1.55 + 1.40 + 0.74 = [5.64]; II: 1.28 + 0.21 + 1.06 + 0.96 + 0.57 = [4.08]; III: 0.70 + 0.15 + 0.53 + 0.47 + 0.34 = [2.19]; IV: 1.17 + 0.19 + 0.89 + 0.70 + 0.38 = [3.33]; Palp: 0.24 -I- 0.07 + 0.13 + (absent) + 0.27 = [0.71]. Epigynum as in Figs. 23, 25, 46, S convex; vulva as in Fig. 40, AD anterior, larger than or equal to HS. Variation: {n = 7). Total length 2.28-3.19; ratios of carapace length/width 1.81-2.07, height/width 0.49-0.56, PER/OQP 2.28-2.80, PER/OAL 2.31-2.93, OQP/OQA 0.94-1.20, diameter AM/PM diameter 1.27-1.60; clypeal height 1.86-2.80 times AM diameter, cheli- ceral length 1.67-2.54 times clypeal height; ratio of sternum length/width 1.14-1.58; ratio GRISWOLD—SCHARFFIA 275 Figures 16-22. — Scharffia chinja new species. 16, 17, 22, Females, from Amani, dorsal view of ab- domen; 18, Male, from from Uzungwa, lateral view; 19, Female, from Mazumbai, dorsal; 20, 21, Female, from Uzungwa; 20, Dorsal; 21, Ventral. of length femur I/carapace width 2.42-3.26. Abdominal shape and markings vary as in male (Figs. 16, 17, 19-22). AD larger than (Fig. 38) or equal to (Fig. 40) HS. Natural history. — The spiders hang be- neath sheet webs in shaded areas in forest (Figs. 2-4). In addition to juveniles and adult females, adult males may be found in intact webs, and both sexes may occur in the same web. 276 THE JOURNAL OF ARACHNOLOGY Figures 23-26. — Scharffia female epigynum and abdominal petiole. 23, 24, Ventral; 25, 26, Lateral; 23, 25, Scharffia chinja new species, from Uzungwa; 24, 26, Scharffia nyasa new species. AT = epigynal atrium, CO = copulatory opening, EF = epigastric furrow, MH = epigynal median hood, ML = epigynal median lobe, S = epigynal scape. (Scale bars for Figs. 23, 25 = 50 jjim, Fig. 24 = 100 |xm. Fig. 26 = 75 fxm.) Distribution. — Eastern Arc mountains and nearby lowlands of Tanzania (Fig. 58). Additional material examined: TANZANIA: Coast Region: Kisarawe District: Kazimzumbwe Forest Reserve, 20 km SW Dar-es-Salaam, 6°57'S,39°03'E, elev. 120-280 m, January-Febru- ary 1991, 1(32$ (Frontier Tanzania Expedition) (ZMUC). Tanga Region: East Usambara Mts. (all C. Griswold, D. Ubick, & N. Scharff, 1995, CAS and ZMUC): Amani, 5°05'S,38°38'E, elev. 950 m, 27 October-9 November, 50c363$; Mbomole Hill, 5°05'S,38°37'E, elev. 1000 m, 5-8 November, 2(315$; Kwamkoro Forest Reserve, 5°10'S,38°35'E, elev. 950 m, 6 November, 8(3 13 $ ; Sangarawe For- est, 38°35'E,5°06'S, elev. 990 m, 5-6 November, 1(33$; Segoma Forest Reserve, 4°58'S,38°45'E, primary rain forest, 17 February 1987, S. Mahunka, T. Poes, & A. Zicsi, 1 $ (HMNH); West Usambara Mts., Mazumbai, 4°49'S, 38°30'E, elev. 1400-1600 m, 10-20 November 1995 (C. Griswold, D. Ubick, & N. Scharff), 15 (345$ (CAS, ZMUC); 1 August 1980, M. Stoltze and N. Scharff, 1(31$ (ZMUC). Morogoro Region: Uzungwa Mts.: Mwanihana For- est Reserve (all N. Scharff, 1984, ZMUC): elev. 500-700 m, 7-16 September, 1 (3; elev. 500-600 m, 11-14 September, pitfalls, 1$; elev. 700 m, 7 Sep- tember, litter, 1 $ ; elev. 1400 m, 27 September, 1 $ ; elev. 1650 m, 25-29 September, litter, 1$; elev. 1800-1850 m, 28-29 September, netted, 1$. Mwanihana Forest Reserve above Sanje (all M. Stoltze & N. Scharff, ZMUC): elev. 600 m, 3 Au- gust 1982, 1$; elev. 700 m, 10 September 1984, 1$; 12 September 1984, netted, 26; elev. 750 m, 1 August 1981, 5(3; elev. 1000 m, 1 August 1981, 2$; 1 August 1982, 1(33$; elev. 1250 m, 25 July 1982, 1(31$; elev. 1650 m, 18 August 1982, litter, 1(32$; pitfall, 3$. Iringa Region: Uzungwa Scarp Forest Reserve above Chita village (all N. Scharff, 1984, ZMUC): elev. 1050 m, 26 October, litter, 1 $ ; elev. 1300 m, 2-6 November, 1 $; elev. 1300 m, 3 GRISWOLD— SC/M/?FF/A 111 Figures 27-29, — Scharffia holmi new species, holotype male, right palpus. 27, Retrolateral; 28, Ventral; 29, Prolateral. A = apical lobe of tegulum, PC = paracymbium, RMP = retromedian cymbial process. (Scale bars for Figs. 27-29 = 50 [xm.) November, litter, 1 $ ; elev. 1400 m, 4 November, netted, 19; 10 November, netted, 29; elev. 1500 m, 9 November, litter. Id; 11 November, netted, ld29 ; elev. 1600 m, 10 November, 1 9 ; elev. 1650 m, 13 November, netted. Id 19. Mbeya Region: Mt. Rungwe SW, elev. 1900 m, 20 August 1984, M. Stoltze & N. Scharff, 1 d (ZMUC). Scharffia holmi new species (Figs. 27-29, 32-36, 39, 58) Types. — ^Male holotype and two female para- types from Mount Elgon, Kenya, elev. 2300 m, 23 December 1937, A. Holm (UUZM). Etymology. — Named in honor of Ake Holm, collector of the type and student of Af- rican montane spiders. Diagnosis. — Distinguished from all Scharf- fia except S. rossi new species by lacking a parembolic process (Figs. 28, 34), having a simple conductor, and having the cephalotho- rax prolonged posteriorly to form a parallel- sided neck (Fig. 32), and from rossi new spe- cies by having the length of the palpal bulb greater than 2.5 X length of median lobe of tegulum (MLT), with the tegulum clearly vis- ible between MLT and embolus (Figs. 28, 34). The epigynum is unique in Scharffia in having a narrow scape (Fig. 33) twice as long as wide, and the vulva unusual in Cyatholipidae in having a lateral afferent duct that is smaller than the spermethecal head (Fig. 39). Description. — Male (holotype): Total length 2.40. Carapace, chelicerae, palpal cox- ae, labium and sternum dark red-brown, un- marked except for dusky maculations along lateral margin of carapace and forming short longitudinal band anteriad of thoracic fovea; ocular area dark gray surrounding AM and be- tween AM and AL, clypeus dark gray in cen- ter from beneath AM to oral margin; coxae, trochanters and legs yellow-white, unmarked except for faint dark mark at base of femur IV; palpi gray-brown, unmarked; abdomen dark gray, dorsum with diffuse longitudinal dark spot in center surrounded by paler cuti- cle. Carapace 1.15 long, 0.54 wide, 0.23 high, greatly prolonged posteriorly to form narrow neck meeting abdomen; PER 0.35 wide, AER 0.34 wide, OAL 0.14; ratio AM:AL:PM:PL, 1.5:1.0:1.12:1.25, PM diameter 0.05. Clypeus 0.15 high, chelicerae 0.25 long. Sternum 0.70 long, 0.46 wide; labium 0.09 long, 0.13 wide; palpal coxae 0.16 long, 0.10 wide. Leg mea- surements (femur + patella + tibia + meta- tarsus + tarsus = [Total]): I: 1.36 + 0.19 + 1.28 + 1.23 + 0.66 = [4.72]; II: 1.02 + 0.17 + 0.83 + 0.76 + 0.49 = [3.25]; III: 0.66 + 0.15 + 0.47 + 0.47 + 0.34 = [2.09]; IV: 0.70 + 0.17 + 0.72 + 0.59 + 0.38 = [2.56]; Palp: 0.23 + 0.07 + 0.07 + (absent) + 0.22 = [0.59]. Palp (Figs. 27-29, 34, 35) with RMP short, blunt, PC broad in lateral view; tegulum apex low, smooth, MLT small and denticulate over median oval area, tegulum exposed be- neath; C simple, single; PP absent. 278 THE JOURNAL OF ARACHNOLOGY Female (paratype): Total length 2.47. Markings and structure as in male (Figs. 32, 36). Carapace 1.20 long, 0.54 wide, 0.26 high; PER 0.35 wide, AER 0.34 wide, OAL 0.14; ratio AM:AL:PM:PL, 1.5:1.37:1.0:1.5, PM di- ameter 0.04. Clypeus 0.11 high, chelicerae 0.27 long. Sternum 0.69 long, 0.45 wide; la- bium 0.10 long, 0.14 wide; palpal coxae 0.19 long, 0.13 wide. Leg measurements (femur + patella + tibia + metatarsus + tarsus = [To- tal]): I: 1.38 + 0.21 + 1.21 + 1.15 + (miss- ing) = [?]; II: 1.02 + 0.18 + 0.85 + 0.76 + 0.49 = [3.28]; III: 0.70 + 0.17 + 0.47 + 0.45 + 0.38 = [2.17]; IV: 0.98 + 0.16 + 0.79 + 0.66 + 0.36 = [2.95]; Palp: 0.21 + 0.08 + 0.10 + (absent) + 0.23 = [0.62]. Epigynum as in Fig. 33, S narrow; vulva as in Fig. 39, AD lateral, smaller than HS. Variation: (n = 2). Total length 2.47-2.72; ratios of carapace length/width 2.25-2.43, height/width 0.49-0.57, PER/OQP 2.36-2.44, PER/OAL 2.54-2.60, OQP/OQA 0.93-0.94, diameter AM/PM 1.50-1.87; clypeal height 2.36-2.44 times AM diameter, cheliceral length 2.06-2.36 times clypeal height; ratio of sternum length/width 1.45-1.53; ratio of length femur I/carapace width 2.55-3.03. Natural history. — Unknown. Distribution. — Known only from the type locality (Fig. 58). Material examined. — Only the type specimens. Scharffia nyasa new species (Figs. 24, 26, 37, 41-43, 47-53, 58) Types. — Male holotype and female para- type from Widdringtonia evergreen forest at 2000 m on Lichenya Plateau on Mt. Mulanje, Malawi, 7 November 1981, R. Jocque (MRAC 156.180). Etymology. — An old name for Malawi. Diagnosis. — Distinguished from all other Scharffia by having the petiole short, length less than 0.17 carapace length (Figs. 24, 41- 43); also leg I extremely long (Fig. 43), femur I of female greater than 3.5, that of male greater than 5.4 times carapace width. Description. — Male (holotype): Total length 2.49. Carapace, palpal coxae, labium and sternum dusky red-brown, chelicerae dark yellow-brown, unmarked except for macula- tions along margin of carapace and anteriad of thoracic fovea; coxae, trochanters and bases of legs yellow-white, legs shading to yellow- brown distally on femora to tarsi, unmarked except segments lighter at joints, palpi yellow- gray, cymbium dark red-brown; abdomen with dorsum black with central longitudinal light band, sides white shading to gray ventrally (Fig. 43). Carapace 0.98 long, 0.62 wide, 0.26 high, not prolonged posteriorly; PER 0.34 wide, AER 0.31 wide, OAL 0.15; ratio AM: AL:PM:PL, 1.2:1.0:1.0:1.1, PM diameter 0.05. Clypeus 0.15 high, chelicerae 0.29 long. Sternum 0.53 long, 0.45 wide; labium 0.10 long, 0.15 wide; palpal coxae 0.18 long, 0.14 wide. Leg measurements (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 3.40 + 0.23 + 3.23 + 3.57 + 1.21 = [11.64]; II: 1.55 + 0.19 + 1.34 + 1.15 + 0.66 = [4.89]; III: 0.83 + 0.17 + 0.62 + 0.57 F 0.40 = [2.59]; IV: 1.38 + 0.19 + 1.06 + 0.85 + 0.47 = [3.95]; Palp: 0.24 + 0.10 + 0.08 + (absent) + 0.28 = [0.70]. Palp (Figs. 48-53) with RMP broadly triangular, PC narrow, sharply angled in lateral view; tegulum apex raised, pustulate, MLT large, with produced trans- verse denticulate ridge; C narrow at base, smooth, with small, narrow secondary pro- cess; PP present, pustulate. Variation: (n = 5). Total length 2.49-3.23; ratios of carapace length/width 1.58-1.73, height/width 0.40-0.48, PER/OQP 2.28-2.71, PER/OAL 2.22-2.29, OQP/OQA 0.93-1.08, diameter AM/PM 1.09-1.50; clypeal height 2.28-2.71 times AM diameter, cheliceral length 1.72-2.00 times clypeal height; ratio of sternum length/width 1.15-1.23; ratio of length femur I/carapace width 5.42-9.48 (Fig. 43). Female (paratype): Total length 2.68. Markings and structure as in male except ab- domen with dorsum dark gray enclosing long median and short anterolateral longitudinal white bands, sides white, venter yellow-gray (Figs. 41-42). Carapace 1.00 long, 0.57 wide, 0.26 high; PER 0.36 wide, AER 0.35 wide, OAL 0.15; ratio AM:AL:PM:PL, 1.3:1. 1:1.0: 1.1, PM diameter 0.05. Clypeus 0.11 high, chelicerae 0.28 long. Sternum 0.55 long, 0.44 wide; labium 0.10 long, 0.17 wide; palpal coxae 0.17 long, 0.13 wide. Leg measure- ments (femur -h patella + tibia + metatarsus + tarsus = [Total]): I: 2.68 + 0.25 + 2.45 + 2.51 + 1.02 = [8.91]; II: 1.51 + 0.21 + 1.19 + 1.06 + 0.62 = [4.58]; III: 0.81 + 0.17 + 0.59 + 0.57 + 0.38 = [2.52]; IV: 1.34 + 0.19 + 1.00 + 0.87 + 0.45 = [3.95]; Palp: 0.21 + GRISWOLD— 5C/M/?FFM 279 Figures 30-36. — Scharffia. 30, 31, Scharffia rossi new species, holotype male, left male palpus; 30, Ventral; 31, Retrolateral; 32-36, Scharffia holmi new species. 32, 33, 36, Paratype female; 32, Dorsal; 33, Epigynum, ventral; 36, Ventral; 34, 35, Holotype male, left male palpus; 34, Ventral; 35, Retrolateral. (Left scale bar for Figs. 30, 31, 33-35, right scale bar for Figs. 32, 36.) 280 THE JOURNAL OF ARACHNOLOGY Figures 37-40. — Scharjfia, cleared female vulvae, dorsal view. 37, Scharffia nyasa new species; 38, Scharffia chinja new species, from Kazimzumbwe; 39, Scharffia holmi new species, paratype; 40, Scharf- fia chinja new species, from Uzungwa. AD = vulval afferent duct, FD = fertilization duct, HS = sper- mathecal head. (Scale bar (Fig. 40, applies to all) = 0.1 mm.) 0.08 + 0.11 + (absent) + 0.27 = [0.67]. Epi- gynum as in Figs. 24, 26, 47, S broad and truncate; vulva as in Fig. 37, AD anterior, larger than HS. Variation: (n = 4). Total length 2.68-3.00; ratios of carapace length/width 1.65-1.76, height/width 0.38-0.46, PER/OQP 2.40-2.46, PER/OAL 2.25-2.43, OQP/OQA 0.93-1.16, diameter AM/PM 1.20-1.40; clypeal height 2.40-2.46 times AM diameter, cheliceral length 2.00-2.45 times clypeal height; ratio of sternum length/width 1.19-1.25; ratio of length femur I/carapace width 3.67-4.67. Natural history. — Data on collection la- bels indicate occurrence in montane forest, where specimens were collected in litter and by sweeping. Distribution. — Known only from the type locality (Fig. 58). Additional material examined. — MALAWI: Mt. Mlanje (all R. Jocque, 1981, MRAC): Thuchila Hut, Nambiti stream, elev. 2000 m, 11 November, 1 (3 1 $ ; Lichenya Plateau, Widdringtonia evergreen forest, elev. 2000 m, 4 November, 3(329, 4-6 No- vember, 19,5 November, 19,7 November, 19,19 November, 1(33 9, 21 November, 8c3309. Scharffia rossi new species (Figs. 1, 30, 31, 54-58) Type . — Male holotype from 1750 m at Naabi, Serengeti Plain, Tanzania, 25 October 1957, E. Ross and R. Leech (CAS). Etymology. — In honor of Edward S. Ross, collector of this and many other new and in- teresting African arthropods. Diagnosis. — Distinguished from all Scharf- fia except S. holmi new species by lacking a parembolic process, having a simple conduc- tor (Fig. 57), and having the carapace pro- longed posteriorly to form a parallel-sided neck (Fig. 1), and from holmi new species by having the median lobe of the tegulum (MLT) large, with bulb length less than 2X length MLT, tegulum nearly hidden between MLT and embolus (Figs. 30, 56). Description. — Male (holotype): Total length 2.66. Carapace, palpal coxae, labium and sternum dark red-brown, unmarked; cox- ae, trochanters, legs and palpi yellow-gray, unmarked except for dark basal annulus on femur IV; abdomen dark gray, venter and sides unmarked, dorsum with yellow-white outlining anteromedian parallel and postero- amSWOLD—SCHARFFIA 281 Figures 41-43. — Scharffia nyasa new species. 41, Female, dorsal; 42, Female, ventral; 43, Male, lateral. lateral converging longitudinal dark marks (Fig. 1). Carapace 1.26 long, 0.61 wide, 0.37 high, greatly prolonged posteriorly to form narrow neck meeting abdomen; PER 0.40 wide, AER 0.39 wide, OAL 0.18; ratio AM: AL:PM:PL, 1.27:1.0:1.09:1.27, PM diameter 0.06. Clypeus 0.18 high, chelicerae 0.27 long. Sternum 0.68 long, 0.59 wide; labium 0.10 long, 0.16 wide; palpal coxae 0.19 long, 0.16 wide. Leg measurements (femur + patella + tibia + metatarsus + tarsus = [Total]): I: 2.15 + 0.23 + 1.98 + 1.81 + 0.81 = [6.98]; II: 1.28 + 0.21 + 1.04 + 0.92 + 0.53 = [3.98]; III: 0.85 + 0.19 + 0.59 + 0.53 + 0.45 - [2.61]; IV: 1.21 + 0.19 + 0.91 + 0.76 + 0.42 = [3.49]; Palp: 0.23 + 0.07 + 0.10 + (absent) + 0.25 = [0.65]. Palp (Figs. 30, 31, 54-57) with RMP short, pointed, PC very broad in lateral view; tegulum apex raised, weakly wrinkled, MLT very large and sparsely den- Figures 44-49. — Scharjfia. 44, 45, Scharffia chinja new species, holotype, left male palpus; 44, Ventral; 45, Retrolateral; 46, Scharffia chinja new species, female, from Uzungwa, epigynum, ventral; 47, Scharf- fia nyasa new species, female, epigynum, ventral; 48, 49, Scharffia nyasa new species, left male palpus; 48, Ventral; 49, Retrolateral. GmSWOhD—SCHAMFFIA 283 Figures 50-53. — Scharjfia myasa new species, right male palpus. 50, Retrolateral; 51, Prolateral; 52, Ventral; 53, Parembolic process. (Scale bars for Figs. 50-52 = 50 jxm. Fig. 53 = 10 |xm.) ticulate over wide median area, tegulum hid- den beneath; C simple, narrow; PP absent. Female: Unknown. Natural history The specimen was col- lected on a hilltop in shade beneath tall um- brella acacias with an understory of grass and stones, either from tree bark or beneath ob- jects on the ground. This dry site was more than 50 km from moist forest (E. Ross, pers. comm.). Distribution.— -Known only from the type locality (Fig. 58). Material examined. — Only the type specimen. DISCUSSION Synapomorphies for Scharffia are the elon- gate sternum (length greater than 1.15X width: Figs. 21, 36) and elongate abdominal petiole. The sternal form is unique within the Cyatholipidae and Synotaxidae. Within these families an annulate anterior abdominal peti- ole (Figs. 11, 26) is uniquely shared with Alaranea Griswold 1997 from Madagascar, and is a synapomorphy uniting these genera: 284 THE JOURNAL OF ARACHNOLOGY Figures 54-57. — Scharffia rossi new species, holotype, right male palpus. 54, Retrolateral; 55, Prola= teral; 56, Ventral; 57, Conductor. (Scale bars for Figs. 54-56 = 50 jam. Fig. 57 = 12.5 |xm.) that of Scharffia is longer than that of Alara- nea, which in turn has a unique dorsal horn (Griswold 1997, figs. 4, 68, 94). Synapomor- phies within Scharffia are the carapace pro- longed posteriorly into a neck uniting holmi new species (Fig. 32) and rossi new species (Fig. 1) and an abdominal petiole longer than 0.24 carapace (Figs. 11, 18) uniting these spe- cies with chinja new species. Are Scharffia components of the Afromon- tane biota (White 1978; Griswold 1991)? Whereas they occur in montane forests of the Eastern Arc mountains and Albertine Rift, they are also recorded from lowland forests and savanna woodland (Fig. 58). Unlike the montane east African Linyphiidae studied by Scharff (1992, 1993), which typically had en- demic species on each mountain within the Eastern Arc, Scharffia chinja new species is widespread. Whether Scharffia are very old (perhaps older than the mountains) and slow to differentiate, or readily dispersed, cannot be easily resolved. Occurrence of Scharffia in lowland forest {chinja) and open, dry country {rossi) suggests that for Scharffia, the Eastern Arc mountains may not be effectively isolated GmSWOhD—SCHARFFIA 285 from one another. On the other hand, the dis- tribution of the sister group of Scharffia (Alar- anea, in Madagascar) is consistent with the Afromontane biogeographic pattern detailed for spiders (Griswold 1991) in which Mada- gascar and the montane forests of eastern Af- rica are sister areas. Several groups of spiders, including Phyxelida and the Lamaika group of the Amaurobiidae Phyxelidinae (Griswold 1990), and Ulwembua and Alaranea plus Scharffia of the Cyatholipidae (Griswold 1997), show this intercontinental disjunction. 286 THE JOURNAL OF ARACHNOLOGY suggesting that their distribution is not the re- sult of accidental dispersal. Their distribution may date from times of former connection or at least greater proximity between Madagascar and eastern Africa, perhaps in the Mesozoic (Rabinowitz et al. 1983). Given the possible great age of this sister-group disjunction, Scharjfia appears to be another component of an ancient Afromontane biota. ACKNOWLEDGMENTS Principal support for this project was pro- vided by National Science Foundation grant DEB-9296271, with additional support from the Exline-Frizzell Fund (California Academy of Sciences), postdoctoral fellowships from the Smithsonian Institution and Kalbfleisch Fellowships from the American Museum of Natural History. The material on which this study was based was made available by Nikolaj Scharff, Zoo- logical Museum, University of Copenhagen (ZMUC), Rudy Jocque of the Musee Royal de UAfrique Centrale, Tervuren (MRAC), Lars Wallin of the Zoological Museum, Uppsala University (UUZM) and Sandor Mahunka of the Hungarian Museum of Natural History, Budapest (HMNH). Additional material is from the collection of the California Academy of Sciences (CAS). Research was made possible through a Research Permit from the Tanzania Com- mission for Science and Technology (COS- TECH) and Residence Permit Class C from the Tanzanian Department of Immigration, and export of specimens made possible by a CITES Exemption Certificate from the Wildlife Division of the United Republic of Tanzania, facilitated by Professor Kim M. Howell of the University of Dar-es-Salaam. Mr. Samwel Y. Fue, Department of Zoology, University of Dar-es-Salaam, served as Sci- entific Counterpart. I thank the following for assistance and hospitality in Dar-es-Salaam: Mr. David Moy- er, Tanzania Coordinator, Centre for Tropical Biodiversity, Dr. Felista Urasa, Head, Depart- ment of Zoology, University of Dar-es-Salaam and Ms. Claire Holliday, Frontier Tanzania. Research in the East Usambaras was made possible by accommodation at the East Usam- bara Conservation and Agricultural Develop- ment Project, Dr. J.K. Ningu, Project Manager, and facilitated by Mr. Massaba I.L. Katigula, East Usambara Catchment Forest Office, Tan- ga, and Mr. Bruno Samuel Mallya, Kwamko- ro. Research in the West Usambaras was made possible by Dr. S.A.O. Chamshama, Dean of Forestry, Sokoine University, Morogoro, and Mr. Modest S. Mrecha, Officer in Charge, Ma- zumbai Forest Reserve. Mr. Lazaro Mbisi, Scan-Tan Tours, is warmly thanked for helping in numerous ways. Nikolaj Scharff and Darrell Ubick col- lected cyatholipids and helped in the field. All habitus illustrations are by Jenny Speckels. Assistance with manuscript preparation was provided by Ms. Johanna Brandriff and Mr. Darrell Ubick; assistance with scanning elec- tron microscopy was provided by Mrs. Susan Breydon (Smithsonian Institution) and D. Ubick (CAS). Nikolaj Scharff took the web photos and Gert Brovad (both ZMUC) made the prints. The manuscript was read and criticized by Norman Platnick and D. Ubick. LITERATURE CITED Brignoli, P.M. 1983. A catalogue of the Araneae described between 1940-1981. Manchester: Manchester Univ. Press, 755 pp. Cambridge, O.P 1903. Descriptions of some new species and characters of three new genera of Araneoidea from South Africa. Annal. S. African Mus., 3:143-165. Coddington, J.A. 1983. A temporary slide mount allowing precise manipulation of small struc- tures. Verb. Naturwiss. Ver. Hamburg (NF), 26: 291-292. Davies, V. Todd. 1978. A new family of spiders (Araneae: Teemenaaridae). Symp. Zool. Soc. London, 42:293-302. Forster, R.R. 1988. Cyatholipidae. Pp. 7-34, In Spiders of New Zealand, vol. 6. Otago Mus. Bull. Forster, R.R., N.I. Platnick & J.A. Coddington. 1990. A proposal and review of the spider fam- ily Synotaxidae (Araneae, Araneoidea), with notes on theridiid interrelationships. Bull. Amer- ican Mus. Nat. Hist., 193:1-116. Griswold, C.E. 1987. A review of the southern Af- rican spiders of the family Cyatholipidae Simon, 1894 (Araneae: Araneomorphae). Annal. Natal Mus., 28:499-542. Griswold, C.E. 1990. A revision and phylogenetic analysis of the spider subfamily Phyxelidinae (Araneae, Amaurobiidae). Bull. American Mus. Nat. Hist., 196:1-206. Griswold, C.E. 1991. Cladistic biogeography of Afromontane spiders. Australian Syst. Bot., 4: 73-89. GRISWOLD— 5C/M^FF/A 287 Griswold, C.E. 1997. The spider family Cyatholi- pidae in Madagascar (Araneae, Araneoidea). J. ArachnoL, 25:53-83. Griswold, C.E., J.A. Coddington, G. Hormiga & N. Scharff. In press. Phylogeny of the orb- web building spiders (Araneae, Orbiculariae: Deino- poidea, Araneoidea). Zool. J. Linn. Soc. Platnick, N.I. 1979. [Review of] Arachnology. New York: Academic Press, 1978, Symp. Zool. Soc. London, 42 (P. Merrett, ed.). Academic Press, New York. Syst. Zool., 28:115-117. Platnick, N.I. 1989. Advances in spider taxonomy: a supplement to Brignoli’s A Catalogue of the Araneae described between 1940 and 1981. Manchester, Manchester Univ. Press, 673 pp. Platnick, N.I. 1993. Advances in spider taxonomy, 1988-1991: with synonymies and transfers 1940-1980. New York Entomol. Soc., 846 pp. Rabinowitz, P.D., M.E Coffin & D. Falvey. 1983. The separation of Madagascar and Africa. Sci- ence, 220:67-69. Roewer, C.F. 1942. Katalog der Araneae von 1758 bis 1940. Bremen: Natura, 1:1-1040. Scharff, N. 1992. The linyphiid fauna of eastern Africa (Araneae, Linyphiidae) -distributional patterns, diversity, and endemism. Biol. J. Linn. Soc., 45:117-154. Scharff, N. 1993. The linyphiid spider fauna (Ara- neae: Linyphiidae) of mountain forests in the Eastern Arc mountains. Pp. 115-132, In Bioge- ography and ecology of the rain forests of eastern Africa. (J.C. Lovett & S.K. Wasser, eds.) Cam- bridge Univ. Press. Simon, E. 1894. Histoire Naturelle des Araignees. 2nd ed. Paris: Roret, 1:489-760. White, E 1978. The Afromontane region. Pp. 132- 143, In Biogeography and Ecology of Southern Africa. (M.J.A. Werger, ed.) Junk, The Hague. Wunderlich, J. 1978. Zur Kenntnis der Cyatholi- pinae Simon 1894 (Arachnida: Araneida: ?Te- tragnathidae). Zool. Beitr. 24:33-41. Wunderlich, J. 1993. Die ersten fossilen Becher- spinnen (Fam. Cyatholipidae) in Baltischem und Bitterfelder Bernstein (Arachnida: Araneae). Mitt. Geol.-Palaont. Inst. Univ. Hamburg, 75: 231-241. Manuscript received I April 1996, revised I Sep- tember 1996. 1997. The Journal of Arachnology 25:288-294 GROWTH RATES IN THE SCORPION PSEUDOUROCTONUS REDDELLI (SCORPIONIDA, VAEJOVIDAE) Christopher A. Brown: Department of Biology, Box 19498, University of Texas at Arlington, Arlington, Texas 76019 USA ABSTRACT. Members of the family Vaejovidae are the dominant species of scorpion in much of western North America, yet relatively little is known of the life histories in this group. In this paper I present data on growth rates of a single litter of Pseudouroctonus reddelli, a troglophilic vaejovid from central Texas. From an initial litter of 53 juveniles, one individual reached instar 8. Nearly 75% of this litter died during instar 2 or 3; this mortality rate was quite high, but consistent with other laboratory studies on vaejovid growth. A comparison with adults from this population suggested that P. reddelli mature at instar 9 for both males and females. Progression factors for two morphological measures during the early instars were more often below the predicted theoretical value of 1.26, while the progression factor for mass was close to the theoretical value of 2.0. No sexual dimorphism in growth rates was observed for instars 2-4. In comparison with other vaejovids, P. reddelli has a larger litter size, shorter instar 1 duration but comparable durations for instars 2-4, and lower morphological progression factors. The scorpion family Vaejovidae Thorell 1876 is a relatively large group, consisting of at least 150 species (Sissom 1990; Stockwell 1992). Recently, Stockwell (1992) revised the Vaejovidae to include only the nominate sub- family (Vaejovinae Thorell 1876), which in- cludes species distributed from southern Can- ada through the United States into central Mexico. Although this family contains prob- ably the most studied species of scorpion [Smeringerus mesaensis (Stahnke 1957); see Polls 1993 and references therein], relatively little is known of the ecology or life history for the majority of vaejovids. For example. Polls & Sissom (1990) give life history data, such as litter size or number of molts to ma- turity, for only 18 species. One of the more interesting vaejovids, both ecologically and taxonomic ally, is Pseudour- octonus reddelli (Gertsch & Soleglad 1972), a relatively large, dark-colored species distrib- uted throughout much of central Texas (Gertsch & Soleglad 1972; Stockwell 1986). As with other vaejovids, individuals may be found under surface debris such as rocks or logs. However, P. reddelli is unusual in that it is troglophilic, with the majority of speci- mens having been captured from caves (Gertsch & Soleglad 1972; Stockwell 1986) despite the lack of any obvious adaptations for cave dwelling (such as lack of eyes or pig- mentation, or elongated appendages) as seen in troglobitic scorpions. Individuals are usu- ally located fairly close to the cave entrance (within the initial 50-100 m). The taxonomic status of P. reddelli remains unsettled. It was initially described by Gertsch & Soleglad (1972) as Vaejovis reddelli before being transferred to a new genus, Pseudour- octonus Stahnke 1974. Subsequent authors (Stockwell 1986; Sissom 1990), noting that characters used to separate Pseudouroctonus from Vaejovis were not unique to either group, returned reddelli to Vaejovis. In the most current treatment, Stockwell (1992) has transferred this species, along with species in the minimus group of Vaejovis, back to Pseu- douroctonus. I have chosen to utilize this most recent nomenclature while acknowledging that there is still debate both to the validity of Pseudouroctonus and the placement of red- delli within a genus. In this paper I analyze growth rates (for both mass and morphometric measures) from a single litter of laboratory-reared P. reddelli. These data represent the first life history in- formation for this species, as well as the first data on changes in mass through ontogeny in a laboratory-reared scorpion. 288 BROWN— GROWTH RATES IN PSEUDOUROCTONUS REDDELLI 289 METHODS Study site. — -The female whose litter was used for this study was collected from Kick- apoo Caverns State Park, located on the bor- der of Kinney and Edwards Counties approx- imately 37 km north of Brackettville, Texas. The park lies on the southwestern Edwards Plateau, a region of limestone hills surround- ing extensive canyons, dominated by Ashe ju- niper {Juniperus ashei) and plateau live oak {Quercus fusiformis) (Lockwood et al. 1993). Underlying the plateau are several under- ground aquifers which have flowed along ma- jor fault lines to create a series of caves and artesian wells (Veni 1988). Within the park, a small population of P. reddelli exists in Kick- apoo Caverns, a series of chambers near the eastern park boundary. From 1992-1994 a to- tal of seven P. reddelli was captured from this cave (bd*, 19). No P. reddelli have been ob- served on the surface at the park, where the dominant scorpion species is Centruroides vit- tatus (Say 1821). Rearing of juveniles.- — On 20 July 1993 a single gravid female P. reddelli was collected from inside Kickapoo Caverns. This female was found after sunset, approximately 40 m from the cavern entrance along a rock out- cropping, by using a portable flashlight with an ultraviolet bulb. Upon return to the labo- ratory the female was weighed (to the nearest gram), then housed in an 18.5 X 7.5 X 9 cm plastic container with a sand substrate (~ 0.5 cm deep). A wet paper towel was provided to serve as a source of moisture and cover ob- ject. The laboratory was maintained on a 14: 10 h light: dark photoperiod at a mean tem- perature of 27.1 °C (range 24-28 °C). The fe- male was fed one adult cricket (Acheta do- mes tied) upon return to the laboratory and was offered one adult cricket every three weeks thereafter until giving birth; she was not fed while carrying offspring. The female gave birth nearly three months after capture, on 10 October 1993. The new- born scorpions (scorplings) oriented them- selves in rows on the female’s back, as is com- monly seen in many species of the genus Vaejovis (e.g., Williams 1969). The juveniles molted into instar 2 after 6 days, and dispersed from the female 6-7 days following molting. Immediately after dispersal, each juvenile was individually weighed to the nearest 0.1 mg. then housed in a 9 cm diameter petri dish con- taining a small square of paper towel. The pe- tri dishes were stacked and kept in a larger (27.5 X 40 X 16 cm) plastic container with paper towels on the bottom which were moist- ened daily. This allowed for the maintenance of adequate humidity without having to wet directly the paper towel in each petri dish (ex- cessive watering can drown immature scorpi- ons and hasten growth of mold on uneaten food). Every third day, the top petri dish in a stack was rotated to the bottom to minimize the effect of any moisture gradient within the box. Every third feeding day I transferred each juvenile to a clean petri dish. Following the molt into instar 4 I added a layer of sand 1-2 mm deep to the petri dish. Rearing oc- curred under the same conditions of temper- ature and photoperiod as described above. Juveniles were maintained on hatchling crickets (one week old or less; mean feeding interval = 6.05 days, range 3-10 days), with the number and/or size of the crickets varying with the scorpion’s instar. In general, I dou- bled either the number of crickets or the size of the cricket offered with each increase in instar. Following the molt into instar 6, each juvenile was moved into a container similar to that which housed the female and fed one adult cricket every three weeks. With the exception of the molt from instar 2 to instar 3 (due to mechanical difficulties with the balance used), juveniles were weighed to the nearest 0. 1 mg following each molt. From the exuvium at each molt (or fol- lowing the death of an individual) I measured three morphological characters [carapace length, metasomal segment V length, and body (prosomal T mesosomal) length] to the nearest 0.01 mm using an American Optical® dissecting microscope equipped with an opti- cal micrometer calibrated at lOX. Body length was computed as the sum of carapace length plus mesosoma length; individual mesosomal segments were not measured separately, as has been recommended by some authors (Stahnke 1970; Sissom et al. 1990). Where possible, I determined sex by looking for the presence of genital papillae, a male secondary sexual characteristic (these were first observed in juveniles during instar 4). For carapace length, metasomal segment V length, and mass I calculated a progression factor (RE) by dividing the value at one instar 290 THE JOURNAL OF ARACHNOLOGY by the corresponding value at the preceding instar (e.g., carapace length at instar 4 divided by carapace length at instar 3). Progression factors were then compared to theoretical val- ues (for mass, RE = 2.0; for length, RE = 1.26, the cube root of 2.0) commonly used to predict the number of molts to maturity in scorpions (reviewed in Erancke & Sissom 1984). Einally, I calculated the instar duration as the number of days between successive molts. All statistical analyses were done using the STATISTICA for Windows (vers. 4.5) computer package (StatSoft 1993). RESULTS AND DISCUSSION The gravid female had an initial mass of 1250 mg when returned to the laboratory and a mass of 778 mg following offspring dis- persal. A total of 53 offspring dispersed from the female; no evidence of cannibalism of ju- veniles was observed either during birth or while the female carried the offspring. This litter size was relatively high for a vaejovid and is over twice the family average of 23 (Polls & Sissom 1990). Only V. spinigerus (Wood 1863), with 66, has a higher reported value (Stahnke 1966). The mean offspring mass following dispersal was 5.8 mg, giving a total litter mass (TLM) of 307 mg. As a measure of reproductive investment by the fe- male I calculated relative clutch mass (RCM) as TLM divided by post-dispersal female mass; this produced a value of 0.395. This represents a lower bound on female invest- ment, as juveniles lose mass while being car- ried by the female (Eormanowicz & Shaffer 1993). This value was below the mean RCMs reported for Centruroides vittatus (0.47-0.53, Eormanowicz & Shaffer 1993; Brown & Eor- manowicz 1995), and for Diplocentrus sp. and V. waueri Gertsch & Soleglad 1972 (0.49 and 0.55, respectively; Brown & Eormanowicz 1996). One individual died following dispersal but prior to the first weighing, leaving 52 juve- niles in the initial sample. Of these, 22 molted into instar 3 (42.3% success rate), 14 molted into instar 4 (26.9% success rate), 10 molted into instar 5 (19.2% success rate), four molted into instar 6 (7.7% success rate), two molted into instar 7 (3.8% success rate), and one molted into instar 8. These success rates are low, but comparable to other studies of vae- jovid post-birth development (e.g., Erancke 1976; Sissom & Erancke 1983; Erancke & Sissom 1984). Death was usually associated with molting. Occasionally this was due to un- known causes, but more often the molting process had begun while a cricket was in the container, and the juvenile had been preyed upon while helpless during emergence from the old exoskeleton. The data for the growth rates from this litter of P. reddelli are summarized in Table 1. The duration of instar 1 (6 days) is shorter than previously reported for any vaejovid, and is less than half the family mean of 12.6 days (Polls & Sissom 1990). The average duration of instars 2-4 is quite consistent at around 100 days, although substantial variability within an instar does exist, more so in instars 2 and 4 than in instar 3. The duration increases dur- ing instars 5 and 6 before decreasing again during instar 7; these values should be regard- ed with some caution because of lower sample sizes. Eor the two comparisons for which I had a reasonable (^10) sample size, the du- ration of time spent in one instar had no effect on the duration of the succeeding instar (Pear- son’s product-moment correlation: instar 2 v.?. instar 3: r = —0.41, P = 0.09, n = 18; instar 3 vs. instar 4: r = 0.29, P = 0.41, n = 10). The durations of instars 2-5 are similar to those reported for Vaejovis bilineatus Pocock 1898 (Sissom & Erancke 1983) and Urocton- us mordax Thorell 1876 (Erancke 1976), but considerably shorter than those for V. coahui- lae Williams 1968 (Erancke & Sissom 1984). It should be noted that differences in rearing and feeding regimes existed between my study and these others, primarily in photope- riod and prey; these may strongly affect the growth rates observed (see below and Polls & Sissom 1990). A bivariate morphometric plot of carapace length versus metasomal segment V length (Pig. 1) showed an overall slight positive al- lometric relationship. In general, the carapace is longer than metasomal segment V for in- stars 2-4 and shorter than metasomal segment V for instars 5-8; this pattern appears to be common in vaejovids (Erancke 1976; Sissom and Erancke 1983; Erancke & Sissom 1984). Within an instar, these two characters were not correlated for instar 2 (r^ = 0.01, P = 0.47, n = 47), but were significantly positively cor- related for instar 3 {P = 0.56, P < 0.001, n = 20), instar 4 (r^ = 0.80, P < 0.001, n = BROWN— GROWTH RATES IN PSEUDOUROCTONUS REDDELLl 291 Table 1. — Growth rates for a single litter of 52 Pseudouroctonus reddeUL Data are given as mean ± SD above, with ranges (sample size n) below. Where there are less than three data points, only the range is given. lestar Mass (mg) Body length (mm) Carapace length (mm) Metasomal segment V length (mm) Duration (days) 2 5.79 ± 0.51 3.98 ± 0.75 1.49 ± 0.05 1.39 ± 0.04 99.6 ± 14.1 43-7.2 (53) 336-635 (44) 1.38-1.57 (47) 1.29-1.48 (50) 74-135 (35) 3 4.28 ± 0.64 1.72 ± 0.07 1.64 ± 0.08 103.1 ± 10.4 3.85-5.95 (17) 1.61-1.84 (20) 1.48-1.80 (21) 87-117 (18) 4 22.5 ± 3.7 538 ± 1.15 1.98 ± 0.17 1.97 ± 0.13 97.7 ± 17.2 17-31 (12) 4.20-7.64 (14) 1.71-2.26 (14) 1.80-2.21 (14) 69-125 (10) 5 44.8 ±11.6 8.12 ± 1.71 2.60 ± 0.20 2.63 ± 0.23 154.5 ± 35.2 29-69 (10) 5.63-10.28 (9) 2.26-2.93 (9) 2.31-3.04 (10) 107-187 (4) 6 94.2 ± 16.0 9.77 ± 1.83 3.13 ± 0.24 330 ± 0.24 73-111.4 (4) 8.13-12.15 (4) 2.90-3.40 (4) 3.04-3.62 (4) 151-192 (2) 7 201-224 (2) 11.7-15.1 (2) 3.82-4.11 (2) 4.01-4.39 (2) 109 (1) 8 434.4 (1) 18.3 (1) 5.16 (1) 6.08 (1) 14) and instar 5 (r^ = 0.91, P < 0.001, n = 9). Carapace length was significantly positive- ly correlated with mass (cube-root trans- formed to equalize dimensionality with length) within iestars 2, 4, and 5 (Figs. 2-4). This relationship is relatively weak initially (r^ = 0.12 for instar 2, w = 47), but becomes quite strong during later instars {P = 0.85 and 0.96, respectively, for instar A {n = 12) and instar 5 (w = 9)]. To examine whether there were differences among sexes in growth rates, I performed a series of r-tests using the 15 individuals (9d, 69) for which I was able to identify sex positively. For instars 2-4, the re- sults (Table 2) showed no dimorphism be- tween the sexes in mass, instar duration or morphometric measures, with the exception of metasomal segment V length in instar 4 (fe- males longer than males). Dimorphism among adults in this population may exist, as all of the adult males captured have been larger than the adult female; however, the sample size of adults is far too small to make any definitive statements, and other authors (Gertsch & So- leglad 1972) have described females larger than males. The mean progression factors (P.F.’s) for carapace length and metasomal segment V length (Table 3) were in general below the theoretical value of 1.26, especially during early development. This was more pro- nounced for carapace length than for meta- somal segment V length. These morphological P.F.’s are less than those previously reported for the Vaejovidae. For carapace length and metasomal segment V length, respectively, average P.F.’s were 1.24 and 1.29 for V. coa- huilae (Francke & Sissom 1984), 1.26 and 1.32 for V. bilineatus (Sissom & Francke 1983), and 1.31 and 1.41 for U. mordax (Francke 1976). As with these three species, the carapace grows less rapidly (P.F.’s were lower) than does the last metasomal segment in all instars of P. reddelii. The mean RE for mass (Table 3) was above the theoretical value of 2.0 in three of the four ratios, although sample size was low in all groups. Mass pro- gression factors have not been reported pre- viously for any vaejovid. The one juvenile to reach instar 8 was a male, as determined by the presence of genital papillae. A comparison of the morphological measurements from this individual (Table 1) to measurements of field-caught P. reddelii males suggests that this may be an immature, and thus males may reach sexual maturity at instar 9. Three adult males captured from Kickapoo Caverns in March 1993 had an av- erage carapace length of 6.45 mm and an av- erage metasomal segment V length of 8.22 mm. Both of these measurements are consid- erably larger than those for the instar 8 indi- vidual (Table 1), and would represent pro- gression factors from instar 8 to instar 9 of 292 THE JOURNAL OF ARACHNOLOGY Figure 1. — Bivariate plot of carapace length (in mm) versus metasomal segment V length (in mm) for a single litter of Pseudouroctonus reddelli. Symbols are as follows: open circles (O) = instar 2; plus signs (+) = instar 3; open diamonds ( 0 ) = instar 4; closed circles (•) = instar 5; asterisks ()K)= instar 6; closed diamonds (♦) = instar 7; crosses (X) = instar 8. The line indicates a 1:1 relationship between the two measures. 1.25 for carapace length and 1.35 for meta- somal segment V length. These P.F.’s are with- in the range of values calculated from earlier instar growth data. Among females from this litter, the two oldest individuals died during instar 5. Using the average RF.’s from instar 6-8 males (1.22 for carapace length, 1.26 for metasomal segment V length), these females would be expected to reach the adult female’s size (carapace length = 6.36 mm, metasomal segment V length = 7.35 mm) at instar 9. Thus, from the laboratory data it appears that maturity in P. reddelli is reached at instar 9 for both sexes. Only two species have pre- Figures 2-4. — Plots of the cube root of mass (in mg’^) versus carapace length (in mm) in Pseudour- octonus reddelli. r represents the Pearson product- moment correlation between the two variables. 2, Instar 2; 3, Instar 4; 4, Instar 5. BROWN— GROWTH RATES IN PSEUDOUROCTONUS REDDELLI 293 Table 2. — Variation between sexes during instars 2-4 in a subset of 15 individual juveniles (9 males, 6 females) from a litter of Pseudouroctonus reddellL Means are reported for each variable. Within each instar, comparisons were made using a t test, ns = not significant. * = significant at P = 0.05. Instar Sex Mass (mg) Carapace length (mm) Metasomal segment V length (mm) Duration (days) 2 male 5.95 1.47 1.4 95 female 5.6 1.5 1.39 97.3 t 1.31 ns 0.97 ns 0.51 ns 0.28 ns 3 male 1.7 1.64 103.7 female 1.73 1.67 99.8 t 0.64 ns 0.88 ns 0.69 ns 4 male 21.4 1.92 1.9 92.4 female 22 1.99 2.02 98 t 0.34 ns 0.73 ns 1.88* 0.45 ns viously been found to require as many instars to reach maturity, both members of the family Diplocentridae Peters 1861: Didymocentrus trinitarius (Franganillo 1930) (9-10 instars; Armas 1982) and Diplocentrus whitei (Ger- vais 1844) (8~9 instars; Francke 1982). For the Vaejovidae the mean instar at maturity is 6.8, with a range of 6-8 (Polis & Sissom 1990). At this point I have no evidence to sug- gest that maturity is reached at different in- stars, either within a sex or between sexes; this phenomenon has been reported for the vaejovid V. coahuilae (Francke & Sissom 1984) as well as a number of species from other families (see Polis & Sissom 1990). Finally, these results should be viewed with caution, for several reasons. First, as has been noted by other authors (reviewed in Polis & Sissom 1990), both environmental factors (e.g., temperature) and feeding history can have an influence on traits such as gestation time and growth rates, such that laboratory studies and field studies of scorpion life his- tories may produce conflicting conclusions for a given species. In this study, this may be the case especially for the estimation of the num- ber of molts to maturity, particularly if pro- gression factors are sensitive to environmental variation (as seems likely). Second, the data presented here represent results from a single litter, so that any genetic variation in repro- ductive investment patterns (e.g., among in- dividuals or populations) was not uncovered. Third, when comparing these results to those Table 3.— Progression factors (P.F.) for carapace length, metasomal segment V length, and mass in Pseudouroctonus reddellL Data are given as mean ± SD above, range (sample size n) below. Where there are less than three data points, only ranges are given. P.F. Carapace length Metasomal segment V length Mass 2 3 1.17 ± 0.05 1.19 ± 0.05 1.09-1.29 (20) 1.10-1.28 (21) 3-^4 1.15 ± 0.07 1.18 ± 0.06 1.00-1.23 (14) 1.11-1.29 (14) 4-^5 1.29 ± 0.03 1.35 ± 0.05 2.03 ± 0.21 1.26-1.34 (9) 1.28-1.45 (10) 1.71-2.35 (9) 5 6 1.22 ± 0.07 1.27 ± 0.04 2.23 ± 0.21 1.13-1.28 (4) 1.23-1.32 (4) 2.04-2.52 (4) 6^7 1.17-1.21 (2) 1.21-1.23 (2) 2.01-2.04 (2) 7 -> 8 1.26 (1) 1.38 (1) 1.94 (1) 294 THE JOURNAL OF ARACHNOLOGY from other studies on scorpion growth, it is important to take strongly into account differ- ences in rearing and feeding conditions, whether the study was field- or lab-based, and whether individuals from various populations were used, since all of these factors are po- tential influences on variation in scorpion growth and reproduction. ACKNOWLEDGMENTS I thank Dan Formanowicz, John Davis, Dan O’Connell and Josh Rose for assistance in col- lecting scorpions. Dave Stuart, superintendent at Kickapoo Caverns, kindly allowed us ac- cess to the cave and all the park around. D. Formanowicz and Norman Johnson kindly re- viewed the initial manuscript; two anonymous reviewers made additional helpful comments. This study was done under permit #14-93 from Texas Parks and Wildlife Department. LITERATURE CITED Armas, L.E de. 1982. Desarrollo postembrionario de Didymocentrus trinitarius (Franganil- lo)(Scorpiones: Diplocentridae). Acad. Sien. Cuba, Misc. ZooL, 16:3-4. Brown, C.A. & D.R. Formanowicz, Jr. 1995. Vari- ation in reproductive investment among and within populations of the scorpion Centruroides vittatus. Oecologia (Berlin), 103:140-147. Brown, C.A. & D.R. Formanowicz, Jr. 1996. Re- productive investment in two species of scorpi- on, Vaejovis waueri (Vaejovidae) and Diplocen- trus linda (Diplocentridae), from west Texas. Ann. Entomol. Soc. America, 89:41-46. Formanowicz, D.R., Jr. & L.R. Shaffer. 1993. Re- productive investment in the scorpion Centru- roides vittatus. Oecologia (Berlin), 94:368-372. Francke, O.F 1976, Observations on the life his- tory of Uroctonus mordax Thorell (Scorpionida, Vaejovidae). Bull. British Arachnol. Soc., 3:254- 260. Francke, O.F. 1982. Birth behavior in Diplocentrus bigbendensis Stahnke (Scorpiones, Diplocentri- dae). J. Arachnol., 10:157-164. Francke, O.F. & W.D. Sissom. 1984. Comparative review of the methods used to determine the number of molts to maturity in scorpions (Arach- nida), with analysis of the post-birth develop- ment of Vaejovis coahuilae Williams (Vaejovi- dae). J. Arachnol., 12:1-20. Gertsch, W.J. & M, Soleglad. 1972. Studies of North American scorpions of the genera Uroc- tonus and Vejovis (Scorpionida, Vejovidae). Bull. American Mus. Nat. Hist., 148:547-608. Lockwood, B., K. Lockwood & M. Lockwood. 1993. Birds of Kickapoo Cavern State Natural Area: a field checklist. Nat. Res. Prog., Texas Parks and Wildlife Dept., Austin, Texas. Polis, G.A. 1993. Scorpions as model vehicles to advance theories of population and community ecology: The role of scorpions in desert com- munities. Mem. Queensland Mus., 33:401-410. Polis, G.A. & W.D. Sissom. 1990. Life history. Pp. 161-223, In The Biology of Scorpions. (G.A. Polis, ed.). Stanford Univ. Press, Stanford, Cali- fornia. Sissom, W.D. 1990. Systematics, biogeography, and paleontology. Pp. 64-160, In The Biology of Scorpions. (G.A. Polis, ed.). Stanford Univ. Press, Stanford, California. Sissom, W.D. & O.F. Francke. 1983. Post-birth de- velopment of Vaejovis bilineatus Pocock (Scor- piones: Vaejovidae). J. Arachnol., 11:69-75. Sissom, W.D., G.A. Polis & D.D. Watt. 1990. Field and laboratory methods. Pp. 445-461, In The Bi- ology of Scorpions. (G.A. Polis, ed.). Stanford Univ. Press, Stanford, California. Stahnke, H.L. 1966, Some aspects of scorpion be- havior. Bull. Southern California Acad. Sci., 65: 65-80. Stahnke, H.L. 1970. Scorpion nomenclature and mensuration. Entomol. News, 81:297-316. Stahnke, H.L. 1974. Revision and keys to the high- er categories of Vejovidae (Scorpionida). J. Ar- achnol., 1:107-141. StatSoft. 1993. STATISTIC A for Windows, ver- sion 4.5. StatSoft, Tulsa, Oklahoma. Stockwell, S.A. 1986. The scorpions of Texas (Arachnida, Scorpiones). Master’s thesis. Texas Tech Univ., Lubbock, Texas. Stockwell, S.A. 1992. Systematic observations on North American Scorpionida with a key and checklist of the families and genera. J. Med. En- tomol., 29:407-422. Veni, G. 1988. The Caves of Bexar County. 2nd ed, Texas Mem. Mus. Speleol. Monogr. 2, Texas Memorial Museum, Univ, of Texas at Austin. Williams, S.C. 1969. Birth activities of some North American scorpions. Proc. California Acad. Sci., 37:1-24. Manuscript received 12 November 1996, accepted 26 May 1997. 1997. The Journal of Arachnology 25:295-306 A COMPARISON OF CAPTURE THREAD AND ARCHITECTURAL FEATURES OF DEINOPOID AND ARANEOID ORB-WEBS Brent D. Opell: Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061 USA ABSTRACT. Orb- webs constructed by the superfamilies Deinopoidea and Araneoidea share a common architecture, but differ in both their orientation and the type of capture thread that they contain. This study uses transformational analyses to determine which web features these clades share and which features are unique to the Araneoidea and may be associated with changes in web orientation and capture thread composition. It examines relationships among spider weight, the cross sectional area of capture thread axial fibers, and features of orb-web architecture in four species of the Family Uloboridae that construct horizontal orb-webs containing cribellar thread and four araneoid species that construct vertical webs containing adhesive capture thread. In both groups, spider weight was positively related to web area and the number of radii in a web were positively related to the number of spirals. In uloborids, weight was negatively related to the number of spirals per web area and axial fiber cross sectional area was positively related to the number of radii per capture spiral turn. In araneoids, spider weight was positively related to axial fiber cross sectional area. The number of radii per capture spiral turn was greater in uloborid webs, and the weight- specific axial fiber cross sectional area was greater in araneoid webs. Many of the features that distinguish araneoid orb-webs appear to equip them to absorb the greater forces of prey strike that are associated with a vertical orb-web orientation. Orb-weaving spiders that produce cribellar capture threads and belong to the superfamily Deinopoidea and those that produce adhesive capture threads and belong to the superfamily Araneoidea share a common web architecture by virtue of their common ancestry (Codding- ton 1986a,b, 1990a, b; Coddington & Levi 1991). The transition from dry, cribellate orb- webs to viscous, adhesive orb-webs is asso- ciated with an increase in species diversity (Bond & Opell pers. obs.) and with changes in web orientation and capture thread com- position that have the potential to alter orb- web architecture and performance. This study uses phylogenetic techniques to determine which web features are shared by both dei- nopoid and araneoid orb-weavers and which features are unique to each clade and may thus reflect differences in the operational dynamics of their webs. It examines associations among spider weight, the cross sectional area of cap- ture thread axial fibers, and orb-web architec- tural features. These relationships provide a better understanding of factors that have con- strained the design and dynamics of spider orb-webs and changes that have been associ- ated with the evolution of araneoid orb- weavers. Deinopoid and araneoid orb-weaving spi- ders are distinguished by differences in web orientation and in the material that covers the axial fibers of their prey capture threads. The horizontal orientation of orb-webs spun by members of the Deinopoidea is plesiomorphic for the Orbiculariae, whereas the vertical ori- entation of orb-webs constructed by the Ara- neoidea is a synapomorphy of this clade (Bond & Opell pers. obs.). As a result of their vertical orientation, araneoid orb-webs tend to intercept faster flying insects and, therefore, are often required to absorb greater forces of impact than are cribellate orb-webs (Craig 1987a; Eberhard 1989). Thousands of fine cri- bellar fibrils surround the axial fibers of cri- bellar capture threads produced by the Deinopoidea (Eberhard 1988; Eberhard & Pe- reira 1993; Opell 1990, 1993, 1994a-c, 1995, 1996; Peters 1983, 1984, 1986, 1992), where- as a chemically complex viscous solution that coalesces into droplets surrounds the homol- ogous axial fibers of adhesive capture threads produced by orb-weaving Araneoidea (Peters 295 296 THE JOURNAL OF ARACHNOLOGY 1995; Tillinghast et aL 1993; Townley et al. 1991; Vollrath 1992; Vollrath et al. 1990; Vollrath & Tillinghast 1991). Each of the droplets of adhesive thread draws in a length of axial fiber that can be played out when ten- sion on the thread increases (Vollrath & Ed- monds 1989). This windlass increases the ex- tensibility of adhesive threads (Kohler & Vollrath 1995) and, thereby, helps maintain web tension and probably reduces capture thread tangling under windy conditions. Differences in architecture can affect orb- web performance. For example, among ara- neoid spiders, orb-webs that have a large number of radii relative to the number of spi- ral turns they contain (radius-rich webs) more effectively stop heavier or faster flying prey than do radius-poor webs (Craig 1987b; Eber- hard 1986). Some of these differences are as- sociated with differences in spider weight. In araneoid orb-weavers, spider weight is direct- ly related to the diameter of the axial fibers within a capture thread (Craig 1987a) and in uloborid orb-weavers, spider weight is direct- ly related to web area and web stickiness (Opel! 1996). Web features such as these have been ex- amined principally among the Araneoidea (e.g., Craig 1987a,b; Eberhard 1986; Risch 1977; Witt et al. 1968) using correlation or regression techniques. Since these studies were done, transformational analysis has be- come the accepted method of analyzing rela- tionships of features in a phylogenetic context (Harvey & Pagel 1991). Therefore, I use this comparative method to examine relationships among spider weight, the cross sectional areas of capture thread axial fibers, and features of web architecture. This analysis found five sig- nificant relationships. Two relationships are shared by both deinopoid and araneoid clades and are hypothesized to be synapomorphies of the Orbiculariae. Two relationships are unique to the Deinopoidea and one relationship is unique to the Araneoidea. Changes in these three relationships are hypothesized to be as- sociated with the origin of araneoid orb-weav- ers. METHODS Species studied. — Ten species of web-spin- ning spiders were studied. Their phylogenetic relationship is shown in Fig. 1. Data for the five araneoid species are taken from the stud- ies of Craig (1987a,b). To these I added data for five species of the family Uloboridae. These species were selected to represent the family’s diversity by including representatives of its major clades (Coddington 1990b) and included the orb-weavers, Waitkera waitak- erensis, Siratoba referena, Uloborus glomo- sus, and Octonoba sinensis and the triangle- web species Hyptiotes cavatus. Hyptiotes ca- vatus was not included in the final comparison of web features, but was used to add resolu- tion to the phylogenetic analysis that gener- ated data used in this comparison. Voucher specimens of each species are deposited in Harvard University’s Museum of Comparative Zoology. Web measurements. — Orb-web architec- ture is sometimes portrayed as being highly stereotypic and species-specific. For example, Foelix (p. 128, 1996) states that “The number of radii varies little within a particular species of orb weaver, and is often characteristic of that species. . . .These numbers of radii imply that many orb weavers show a species-specific geometry in their webs. It is thus often pos- sible to identify a certain spider solely by its characteristic web structure.” However, while acknowledging that “in a local fauna species of orbweavers can often be determined from their webs”, Eberhard (p. 342, 1990) docu- ments a number of factors that contribute to intraspecific differences in orb-web architec- ture and cautions that: “The impression of species-specificity may usually, however, be the product of lack of information. . . . Given the long-standing and repeated documentation of substantial intraspecific variation in at least gross web characteristics such as numbers of radii, spiral loops, spacing between loops, an- gle of web plane with the vertical, web area, top-bottom asymmetry, and stabilimenta, Levi’s prediction that species- specificity will be uncommon seems likely to be correct.” This study attempts to minimize the poten- tial problem of intraspecific variation in ulo- borid web features by using species means in transformational analyses. As only one web per individual was measured, it does not ad- dress the intra-individual variability in web features that may result from differences in nutritional levels or reproductive status. How- ever, the minimum sum of squares algorithm used by the transformational analysis mini- mizes the hypothesized evolutionary changes OPELL— ORB WEB FEATURES 297 DEINOPOBDEA -- Uloboridae Waitkera waitakemnsis (Chamberlain, 1946) 1 r Siratoba referena (Muma & Gertsch, 1964) Uloborus glomosus (Waickenaer, 1841) Octonoba sinensis (Simon, 1880) Hyptiotes cavatus (Hente, 1847) ARANEOIDEA — Araneidae — Mangora pia Chamberline & Ivie, 1936 Cyclosa caroli (Hentz, 1850) Micrathena schreibersi (Perty, 1833) Tetragnathidae Leucauge giobosa (O. Pickard-Cambridge, 1889) Theridiosomatidae Epilineutes globosus (O. Pickard-Cambridge, 1896) Figure 1. — Cladogram of the species included in this study (from Coddington 1990b; Coddington & Levi 1991; Levi 1985). I T1 that are used in statistical analyses and, there- by, makes the results of these analyses con- servative. I dusted the webs produced by adult fe- males with com starch to make their threads more visible (Carico 1977) and photographed only webs that were not damaged. I photo- graphed the webs of W. waitakerensis in the field and those of O. sinensis in a greenhouse. In field photographs of the other three species, it was difficult to distinguish the threads from background vegetation. Therefore, I allowed these spiders to build their webs in individual plastic boxes that contained a framework of wooden dowel rods and photographed these webs against a black background. Spiders were placed into boxes immediately after be- ing collected and were not fed. Those that did not build a web within three days of capture were released. Boxes were housed in an en- vironmental chamber with a 1 h dawn, 11 h light, and 1 h dusk light cycle. Temperature was maintained at 24 °C and relative humidity ranged from 80% during the night to 70% at dusk and dawn, to 60% during the day. After photographing a spider’s web, I col- lected and weighed to the nearest 0.00 mg the spider that produced it. Conspecific web cap- ture has not been studied in the species that were included in this study. Therefore, there is a small possibility that some spiders whose webs were photographed may not have actu- ally constmcted the webs in which they were found. It is possible that the boxes in which spiders constmcted their webs may have affected the 298 THE JOURNAL OF ARACHNOLOGY size of these webs. To assess this for the two orb-weaving species, I determined the mean framework diameter and the maximum cap- ture spiral diameter for each species and com- pared this with the dimensions of the plastic boxes in which these spider’s constructed their webs. Mean framework diameter was com- puted from the minimum and maximum lengths of straight lines that extended across a web’s center to its outermost, non-sticky framework threads. Maximum capture spiral diameter is the length of a straight line ex- tending across a web’s center to its outermost capture thread. The dowel rods inside these boxes formed a frame with inner dimensions of 29.0 X 21.5 cm, although some spiders at- tached their threads to the walls of boxes and, therefore, perceived the space available for web construction as 1--2 cm greater than this. The mean web diameters for S. referena was 11.7 cm {n = 26, SE = 0.7) and that for U. glomosus was 18.7 cm {n = 29, SE — 0.5). The maximum capture spiral diameter for S. referena was 11.8 cm {n = 26, SE = 0.7) and for U. glomosus was 19.1 cm {n = 29, SE = 0.5). Thus, the mean web diameter of S. re- ferena is 54% and the maximum capture spiral diameter is 54% of the minimum box dimen- sion. For U. glomosus these values are 87% and 89%, respectively. Therefore, box size clearly did not restrict the size of S. referena webs and probably did not cause U. glomosus to construct smaller webs than those found in the field. To assess the effect of box size on the tri- angle-webs of H. cavatus, I compared the length of the second radius (the top of the web being the first of four radii) in webs construct- ed by adult females in the field with those constructed in the laboratory. The length of this radius is highly correlated to other web parameters (Opell unpubl. data) and is, there- fore, a good index of web size. The mean length of the second radius in webs construct- ed in the field was 11.1 cm {n ^ 19, SE = 0.7) and that for webs constructed in the lab- oratory was 14.3 cm {n = 30, SE = 0.5). The values for both populations were normally distributed (Shapiro-Wilk W-statistic P > 0.45) and a Utest showed that their means were different {t = 3.928, P < 0.001). This indicates that the structural spacing that H. ca- vatus typically encounters in the field limits the size of its web to a greater degree than that provided in the laboratory. Consequently, webs constructed in the laboratory may be considered to be of optimal size. From enlarged photographic prints I count- ed the number of radii and spirals in each web and measured the web’s area with a digitizing tablet. There are two measurements of web area that can be taken: total area, the area in- scribed by a web’s frame lines, and capture area, the area between a web’s outermost and innermost capture spirals. I measured total web area because it seemed a better index of web size for comparisons of web architecture, whereas capture area seems more appropriate for assessing a web’s prey capture potential. Craig (1987b) does not report total web area, but does give the mean radius length for each species she studied. I used these data to com- pute the total area of each species’ web as if it were a circle. Thread diameters. — The diameters of ulo- borid axial fibers are taken from table 2 in Opell (1996). For cribellar threads, two axial fibers were measured per web. This approach assumes that axial fiber diameter is uniform within a web and does not address the possi- bility that axial fiber diameter changes during the course of capture spiral production. The cribellar fibrils of these threads help hold their axial fibers apart, allowing the diameter of a single fiber to be measured under a transmis- sion electron microscope. The total axial fiber cross sectional area of these cribellar threads was computed as the sum of the cross sec- tional areas of their two fibers. In contrast, even the interdroplet regions of araneoid cap- ture threads are coated by a thin layer of vis- cous material. Although the water in this ma- terial evaporates under the high vacuum of an electron microscope, it leaves a thin, electron- dense residue that coats the axial fibers, mak- ing them appear as a single strand, whose in- dividual fibers cannot be distinguished under either the scanning or transmission electron microscope (Craig 1987b; Opell unpubl. obs.). I computed the combined cross sectional areas of the axial fibers of these threads as the sum of the areas of two circles, each with a di- ameter of half the capture thread diameter re- ported by Craig (1987b). This provides a more appropriate estimate of axial fiber cross sec- tional area than treating the contiguous fibers as if they were a single fiber. Statistical analysis.— The relationships OPELL— ORB WEB FEATURES 299 among spider weight, thread diameter, and web features cannot be determined using tra- ditional regression techniques, as the species included in this study are evolutionarily relat- ed, and their values are not strictly indepen- dent (Harvey & Pagel 1991). To minimize the effect of phylogenetic position, I employed Huey and Bennett’s (1986) method for eval- uating the relationships among continuous variables whose states are hypothesized to be functionally linked. This method has three steps: 1) the state of each character in a tax- on’s most immediate hypothetical ancestor is determined, 2) the change from this ancestral state to the state expressed by extant members is computed for each character, and 3) the re- lationship between these changes in character states are evaluated by Pearson correlation. If this analysis shows that changes (either posi- tive or negative) in two characters are corre- lated, then their states can be considered to have coevolved. I determined ancestral values for uloborids and araneoids separately using the unrooted, minimized sum of squared changes option in the continuous character tracing section of the MacClade 3.02 phylo- genetic program (Maddison & Maddison 1992). Although this study compares the only uloborids that construct horizontal orb-webs and araneoids that construct vertical orb-webs, H. cavatus and L. globosus were included in determinations of ancestral values to increase the resolution of these computations. As spider weight has the potential to affect web features, I used a one way ANOVA to determine if weight differed between: 1) ulo- borid and araneoid species, 2) uloborids that spin horizontal webs and araneoids that spin vertical webs, 3) uloborid and araneoid spe- cies that produce horizontal webs and those that produce vertical webs. These tests showed no differences in spider weight that would compromise this study’s findings (f — 0.95-L12, P = 0.32-0.36). RESULTS Values and their normality.— Because all values or their natural logs are normally dis- tributed, parametric statistics were used in their analysis* Tables 1 and 2 give values for uloborid and araneoid species and Table 3 pre- sents the ancestral values used in transfor- mational analyses. A Shapiro-Wilk W-statistic test of normality showed that changes in axial fiber cross sectional area, the number of radii per web area, the number of spirals per web area, and in the number of radii per spiral turn were normally distributed {P > 0.28) for both the four uloborid species that constructed hor- izontal webs and the four araneoid species that constructed vertical webs. Change in spider weight was not normally distributed {P = 0.002) and the normality of change in web area was questionable {P ^ 0.059). However, changes in the natural logs of these two latter values were normally distributed {P > 0.48) for both the four orb-weaving uloborids and the four araneoid orb-weavers that constructed horizontal orb- webs. The number of radii per spiral turn and the weight- specific cross sec- tional area of axial fibers (Table 5) were nor- mally distributed (Shapiro-Wilk W-statistic P > 0.30) for both deinopoid and araneoid clades. Correlation between features.— Five fea- tures were shown by Pearson correlation to be significantly correlated for at least one group of orb- weavers (Table 4). Given the small sample size for each clade of spiders and the high correlation values obtained in these anal- yses, I accept as significant correlations with P < 0.10. For both cribellate and adhesive orb-weavers spider weight and web area are positively correlated, as are the number of ra- dii per web area and the number of spirals per web area. However, the transition from cri- bellate to adhesive orbs appears to have been associated with three changes: 1) the gain of a positive relationship between spider weight and capture thread cross sectional area, 2) the loss of a negative relationship between spider weight and the number of spirals per web area, and 3) the loss of a positive relationship between web cross sectional area and the number of radii per spiral turn. Differences between uloborid and ara- neoid orb-webs. — Both the number of radii per spiral turn and the weight- specific cross sectional area of axial fibers differed between uloborids with horizontal orb-webs and ara- neoids with vertical orb-webs (Table 5). The number of radii per spiral turn was greater in uloborid orb-webs and the weight- specific ax- ial fiber cross sectional area was greater in araneoid orb-webs. DISCUSSION Common web features. — Two architectur- al relationships appear to be plesiomorphic for 300 THE JOURNAL OF ARACHNOLOGY Table 1. — Features of the webs and threads of five species of Uloboridae. Mean ± 2 standard errors, sample size. Weights are those of individuals whose web features were measured. * Although many webs constructed by these two species are essentially horizontal, some are constructed at angles of up to about 45°. Waitkera waitakerensis Siratoba referena Uloborus glomosus Octonoba sinensis Hyptiotes cavatus Web orientation horizontal horizontal* horizontal horizontal* vertical Weight (mg) 7.84 ± 0.70 3.93 ± 0.44 7.16 ± 0.76 12.16 ± 1.58 8.41 ± 1.22 n = 32 n = 26 n = 29 n = 24 n = 30 Axial fiber: diameter (nm) 236 ± 44 292 ± 70 307 ± 46 340 ± 30 419 ± 22 n = 6 n = 5 n = 5 n = 11 n = 11 cross sectional 0.09 ± 0.04 0.14 ± 0.06 0.15 ± 0.04 0.19 ± 0.04 0.28 ± 0.02 area X 2 ([xm^) n — 6 n = 5 n = 5 n = 11 n = 15 weight-specific area 19 ± 28 31 ± 13 21 ± 6 15 ± 4 40 ± 6 (ixmVmg X 10"3) n = 6 n = 3 n = 5 n = 16 n = 15 web area (cm^) 111 ± 28 109 ± 26 289 ± 32 642 ± 100 155 ± 16 « = 32 n = 26 n = 29 n = 24 n = 30 Radii: length 4.0 ± 0.4 3.0 ± 0.5 6.3 ± 0.5 8.8 ± 0.8 — n = 25 n = 23 n ^ 21 n = 24 — number 21 ±2 35 ±2 34 ± 2 50 ± 2 4 ± 0 n = 32 n = 26 n = 29 n = 24 n = 30 number/area 0.17 ± 0.02 0.42 ± 0.08 0.12 ± 0.02 0.9 ± 0.02 0.33 ± 0.02 n = 32 n = 26 n = 29 n = 24 n = 30 Capture thread spirals: number 12 ± 2 14 ± 2 13 ± 2 17 ± 2 16 ± 2 n = 32 n = 26 n = 29 n = 24 n = 30 number/area 0.07 ± 0.00 0.15 ± 0.02 0.5 ± 0.00 0.3 ± 0.00 0.11 ± 0.02 n = 32 n = 26 n = 29 n = 24 n = 30 radii/spiral turn 2.37 ± 0.14 2.64 ± 0.20 2.77 ± 0.22 2.98 ± 0.24 0.12 ± 0.02 n = 32 /I = 26 n = 29 n = 24 n = 30 orb-weaving spiders by virtue of their pres- ence in both deinopoid and araneoid orb- weavers. Both clades exhibit a positive rela- tionships between spider weight and web area and between the number of radii per web area and the number of spirals per web area (Table 4). The positive relationship between spider weight and web area in both clades shows that this relationship plays an important role in the foraging dynamics of orb- weaving spiders. As spider weight is directly related to metabolic rate (Anderson & Prestwich 1982) and as web size is directly related to prey capture (Brown 1981), this relationship indicates that an orb- web’s ability to capture prey tends to scale to a spider’s metabolic needs. However, it is im- portant to note that other factors may also af- fect web performance. These include prey availability in the microhabitat where a web is placed (Riechert & Cady 1983; Wise & Barata 1983; Craig et al. 1994), web and spi- der visibility to insects (Craig 1988, 1990; Craig & Bernard 1990; Craig & Ebert 1994; Craig & Freeman 1991), web orientation (Chacon & Eberhard 1980; Eberhard 1989), the ability of a web to absorb the force of an insect strike, web stickiness (Craig 1987b; Eberhard 1986, 1989), spider response time (Eberhard 1989), and the presence of other orb- weaving species (Spiller 1984). An interspecific comparison of uloborid species (Opell 1996) showed a direct relation- ship between spider weight and web area, but OPELL— ORB-WEB FEATURES 301 Table 2. — Features of the webs and threads of five araneoid species, as given by Craig (1987a, b). Mean ± 2 standard errors, sample size. *No variance was provided, as indices were computed from species means. Mangora pia Cyclosa caroli Micrathena schreibersi Leucauge globosa Epilineutes globosus Web orientation vertical vertical vertical horizontal vertical Weight (mg) 21.2 5.3 146 2.7 0.8 Axial fibers of capture thread: diameter (nm) 1900 1038 3040 760 350 cross sectional area (jxm^) 1.42 0.42 3.63 0.23 0.05 weight- specific area (gmVmg X 10" 9 67 79 25 85 63 Web area (cm^) 216 ± 0.7 150 ± 0.5 347 ± 1.9 125 ± 0.7 61 ± 1.7 n = 42 n = 22 n = 16 n = 16 n = 33 Radius length (cm) 8.3 ± 0.68 6.9 ± 0.56 10.5 ± 1.10 6.3 ± 0.66 4.4 ± 1.14 li n = 22 « = 16 n= 16 n = 33 Radii/web area 0.24* 0.32 ± 0.06 n = 22 0.13 ± 0.02 n = 16 0.15 ± 0.02 n= 16 0.06* Capture spirals/web area 0.25* 0.30 ± 0.20 n = 10 0.09 ± 0.02 n = 9 0.18 ± 0.04 n = 8 0.10* Radii/spiral turn 2.20* 1.10* 1.40* 0.83* 0.11* studies of araneoids lead to contradictory con- clusions. The latter situation may be explained by the fact that these studies are a mix of in- traspecific and interspecific comparisons and that the results of interspecific studies were not analyzed in a phylogenetic context. Risch (1977) measured the weights and spiral areas (area encompassed by the web’s inner- and outer-most spiral turn) of juveniles and adult females of four araneid species. His data do not show a strong relationship between these variables, although the species he studied were more similar in weight than the araneoid species included in the current study. Several intraspecific comparisons of the size of adult female araneoids and the size of their webs show that larger or heavier spiders tend to construct larger webs (Eberhard 1988; Witt et al. 1968), one found no such association in two species (Brown 1981), and another found that adding weights to adults reduced the length of thread in their webs (Christensen et al. 1962). These studies and those of the effect of silk supply, and, by implication, spider nu- trition, on web size (Eberhard 1988) demon- strate that web size is plastic and document some of the proximate factors that influence this parameter. Phylogenetic comparisons, like those presented in this study, provide a com- plementary perspective by documenting ulti- mate factors that influence orb-web architec- ture. Table 3. — Ancestral values used in transformational analyses. The position of these six nodes is given in Figure 1. Node 1 Node 2 Node 3 Node 4 Node 5 Node 6 Weight (mg) Total area of 7.83 9.76 9.96 33.4 61.6 6.9 axial fibers (gm^) 0.21 0.21 0.29 1.29 1.78 0.31 Web area (cm^) 160 396 248 208 235 120 Radius length (cm) 4.3 7.0 — 7.7 8.4 5.7 Radii/area 0.28 0.15 0.24 0.22 0.22 0.14 Spirals/area 0.11 0.06 0.09 0.14 0.18 0.14 Radii/spiral 2.23 2.38 1.40 1.68 1.39 0.96 302 THE JOURNAL OF ARACHNOLOGY Table 4. — Comparison of the relationships found among four cribellate orb-weaving species and four adhesive orb-weaving species using Pearson correlation. Significant values (P < 0.10) are indicated by an asterisk (*). Horizontal, cribellate orb-webs Vertical, adhesive orb-webs Change in L^ weight and in L^ web area r = 0.94* P = 0.056* r = 0.96* P = 0.038* Change in radii per web area and in spirals per web area r = 0.99* P = 0.007* r = 0.91* P = 0.092* Change in L^ weight and in spirals per web area r = -0.92* P = 0.078* r = -0.75 P = 0.254 Change in axial fiber cross sectional area and in radii per spiral turn r = 1.00* P = 0.001* r = 0.37 P = 0.634 Change in L^ weight and in L^ axial fiber cross sec- tional area r = 0.16 P = 0.844 r = 0.94* P = 0.062* The positive relationship between the num- ber of radii per web area and the number of spirals per web area factors out web area and, therefore, reflects a positive relationship be- tween the number of radii and the number of spirals in a web. This relationship has been noted by Eberhard (1972, 1986), who con- cluded that, although there are exceptions, the number of radii and spiral turns “are about equal”. Although the current study is based on only nine orb-weaving species, it suggests that orb-webs tend to have more radii than spirals. The webs of the nine orb-weaving species studied had a mean radii per spiral turn ratio of 1.88. However, this study includ- ed four species of the family Uloboridae, a group that Eberhard (1986) considers to have a greater than typical number of radii. When these uloborid species are excluded, the mean ratio drops to 1.23 radii per spiral turn. Among araneoids, the number of radii de- crease as spiders develop (Risch 1977; Wiehle 1927; Witt et al. 1968). This may indicate that larger araneoid species tend to have fewer ra- dii in their webs than do smaller species. However, as only one very large araneoid spe- cies was included in the current study, size alone cannot account for the lower radii per spiral turn ratio in araneoid webs (Table 5). Differences in cribellate to adhesive orb- web. — The evolution of the Araneoidea was associated with a shift from horizontal orb- webs that contained cribellar capture threads to vertical orb-webs that contained adhesive capture threads (Bond & Opell pers. obs.). The vertical orientation of araneoid orb-webs subjects them to greater forces of prey impact than does the horizontal orientation of ulo- borid orb-webs (Craig 1987a; Eberhard 1989). This kinetic energy is absorbed in two major ways: some is borne by the web’s radii and frame threads and some is dissipated by aero- dynamic damping as the web extends and its capture threads resist movement through the air (Lin et al. 1995). The greater weight-specific cross sectional Table 5. — Comparison of two web features in uloborid and araneoid orb- webs. Mean ± 2 standard errors. Below the name of each index appears the results of a Ltest. Uloborid species Araneoid species with horizontal with vertical orb-webs orb-webs {n = 4) {n = 4) Radii/spiral turn (t = 3.30, P = 0.016) Weight- specific axial fiber cross sectional area 2.69 ± 0.26 1.20 ± 0.86 p^mVmg X 10-3 ^ 3 04, P = 0.023) 22 ±1 59 ± 23 OPELL— ORB-WEB FEATURES 303 areas of araneoid axial fibers (Table 5) indi- cate that these adhesive capture threads are stronger than those constructed by uloborids and, thus, better adapted to transfer greater forces to the web’s stronger radial threads. For the five araneoid species, there is a positive relationship (Pearson r = 0.97, P = 0.007) between the total axial fiber cross sectional area computed in this study and the capture thread tensile strength reported by Craig (1987a). As the axial fibers of cribellar and adhesive capture threads are homologous, the cross sectional area of axial fibers in cribellar thread is probably also a good index of thread tensile strength. Although the spectral prop- erties of light reflected by cribellar and ad- hesive threads differ (Craig & Bernard 1990), these measurements include the non-homolo- gous cribellar fibril and adhesive material that covers the axial fibers. Therefore, these dif- ferences do not necessarily show that the pro- tein composition of the axial fibers of these threads differs. Architectural differences between uloborid and araneoid orb- webs suggest that their func- tional dynamics also differ. Radius-rich webs, like those constructed by uloborids (Table 5), tend to be stiff and radius-poor webs, like those constructed by araneoids, tend to be more extensible (Craig 1987b). The study of Lin et al. (1995) suggests that the more ex- tensible a web is, the more kinetic energy it is able to dissipate through aerodynamic dampening. Therefore, the greater extensibil- ity of adhesive capture threads (VolLrath & Edmonds 1989; Kohler and Vollrath 1995) may enhance aerodynamic dampening by in- creasing overall web extensibility. The greater extensibility of adhesive capture thread may also serve to dissipate some force in the im- mediate area of a prey strike before transfer- ring the remanding force to adjacent threads. Additional evidence that the replacement of cribellar threads by adhesive threads changes web dynamics comes from a comparison of vertical and horizontal adhesive orb-webs. If differences between uloborid and araneoid orb-webs are associated principally with dif- ferences in web orientation, then they should also be observed when horizontal and vertical adhesive orb-webs are compared. However, Craig’s (1987a) data suggest that horizontal araneoid orb-webs have fewer, not more, radii per spiral turns than vertical araneoid orb- webs. This is contrary to the difference be- tween horizontal uloborid and vertical ara- neoid orb-webs observed in this study and suggests that the replacement of cribellar threads by adhesive threads may also enhance a web’s ability to dissipate the force of a prey strike. It may also indicate that the radial threads of araneoid orb-webs are stronger than those of deinopoid orb-webs, either because they have greater diameters or different silk composition. The lower radius-to-capture- spiral ratio of araneoid orb-webs may also contribute to the positive relationship between spider weight and axial fiber cross sectional area that char- acterizes vertical araneoid orb- webs (Table 4; Craig 1987a). As capture threads become a more prominent component of the vertical ar- aneoid orb-web, they play a greater role in transferring force to the web’s stiffer radial threads (Lin et al. 1995) and must be strong enough to withstand this force. However, there appears to be a limit on the amount of material that an orb-weaving spider can devote to cap- ture thread production (Eberhard 1972, 1989; Peters 1937; Witt et al. 1968). As the total volume of adhesive capture thread in a spi- der’s web is directly related to spider weight (Opell unpubl. obs.), axial fiber diameter may ultimately be determined by the competing re- quirements that a spider must produce a length of capture thread that is long enough and sticky enough to capture sufficient prey and strong enough to withstand the force of prey impact. As a spider’s weight affects both its metabolic demand (Anderson & Prestwich 1982) and total thread volume, it is not sur- prising that the cross sectional area of ara- neoid axial fibers is related to spider weight. In uloborids, the cross sectional area of cap- ture thread axial fibers is not related to spider weight, but instead to the maximum distance that a capture thread spans in the web (Opell 1994d). This difference and the smaller weight-specific axial fiber cross sectional ar- eas of cribellar threads suggest that different factors influence axial fibers of uloborids and araneoids. The lower forces of prey impact that uloborid webs typically experience may not require the axial fibers of their capture threads to be as strong. Additionally, as in ar- aneoids, the volume of material that these spi- ders can devote to capture thread production appears to be limited (Eberhard 1972). There- 304 THE JOURNAL OF ARACHNOLOGY fore the large amount of silk volume that ulo- borids must devote to the cribellar fibrils of their capture threads to achieve thread sticki- ness (Opell 1994b, 1996) may indirectly re- strict that amount of silk that can be expended as axial fibers. Cribellate orb-webs are characterized by a negative relationship between spider weight and the number of spirals per web area and by a positive relationship between axial fiber cross sectional area and the number of radii per spiral turn. Neither relationship is present in araneoids. The first relationship indicates that spiral spacing increases as spider size in- creases. Spiral spacing may be more highly constrained in uloborids because the webs that these spiders construct appear to be less well equipped than araneoid orb-webs to retain prey and because uloborids are less well equipped than araneoids to subdue intercepted prey. Not only do horizontal webs retain prey for shorter periods of time than vertical webs with threads of the same stickiness (Eberhard 1989); but, relative to the weight of the spider that produced them, cribellar capture threads are less sticky than adhesive threads (Opell unpubl. obs.). Additionally, uloborids lack poison glands and must rely entirely on silk wrapping to quiet prey and prevent their es- cape from the web (Lubin 1986; Opell 1979). As orb-webs trap prey more efficiently when capture spiral spacing exceeds prey diameter (Chacon & Eberhard 1980; Eberhard 1986), the more closely spaced spirals of orb-webs constructed by small uloborid species may be particularly important in equipping these webs to intercept prey that they can retain and that spiders can subdue. The greater stickiness of adhesive capture threads (Opell unpubl. obs.) may further increase the prey capture efficien- cy of vertical araneoid orb-webs and allow their spiral spacing more latitude to differ in ways that adapt webs to a particular habitat or prey type. In uloborids, but not in araneoids, axial fi- ber cross sectional area increases as the num- ber of radii per spiral turn increases. In view of the positive relationship between the num- ber of radii per web area and the number of spiral turns per web area, this indicates that increased axial fiber cross sectional area adds to overall web strength by complementing an increase in the number of radii rather than by compensating for a decrease in the relative number of spiral turns. The lower extensibility of cribellar threads (Kohler & Vollrath 1995) and the stiffer, radius-rich webs of uloborids (Table 5) may help explain why the cross sec- tional areas of their capture threads responds to this change in web architecture and those of araneoids do not. If uloborid orb-webs have a lower ability to dissipate force through ex- tension and aerial dampening, they may meet this challenge by becoming stronger. If the ax- ial fibers’ chemical structure is unchanged, then increased strength is gained by increased cross sectional area. Conclusions. — Orb-webs constructed by members of the deinopoid and araneoid clades share many features, including an area that is related to spider weight. However, this study shows that there are important architectural differences between the webs that are spun by members of these sister clades. The functional implications of these differences are consis- tent with the observation that vertical araneoid orb-webs typically experience greater forces of prey impact than do deinopoid orb- webs. Compared to horizontal orb-webs, the vertical orb-webs of araneoids appear to have stronger capture thread axial fibers and to be better equipped to implement aerodynamic damp- ening by virtue of their lower radius-to-spiral ratio. The greater extensibility of adhesive capture thread may contribute in a minor way to overall web extensibility and force dissi- pation, but the model of orb-web dynamics developed by Lin et al. (1995) suggest that it does not play a major role. Therefore, the se- lective advantage of adhesive capture thread over cribellar capture thread may be due prin- cipally to the greater economy and greater stickiness of adhesive thread (Opell unpubl. data) and to its reduced ultra violet reflectance (Craig & Bernard 1990) that makes it less vis- ible to insects and allows araneoid orb-weav- ers to occupy an expanded range of micro- habitats (Craig et al. 1994). ACKNOWLEDGMENTS I thank Catherine Craig and anonymous re- viewers for providing useful comments on this manuscript. Field studies were conducted at the Center for Energy and Environment Sci- ence’s El Verde field station in Puerto Rico, the Organization for Tropical Studies’ La Sel- va field station in Costa Rica, and the Amer- ican Museum of Natural History’s Southwest- OPELL— ORB-WEB FEATURES 305 era Research Station in Arizona. The Auckland Regional Council Parks Committee, the New Zealand Department of Conserva- tion, Northland Conservancy Office, and the Whangarei District Council granted collecting permits for studies of Waitkera waitakerensis. This material is based upon work supported by the National Science Foundation under grants BSR-8917935 and IBN-9417803. LITERATURE CITED Anderson, J.E & K.N. Prestwich. 1982. Respira- tory gas exchange in spiders. Phys. ZooL, 55:72- 90. Brown, K. 1981. Foraging ecology and niche par- titioning in orb- weaving spiders. Oecologia, 50: 380-385. Carico, J.E. 1977. A simple device for coating orb webs for field photography. Bull. British Arach- nol. Soc., 4:100. Chacon, P. & W.G. Eberhard. 1980. Factors af- fecting numbers and kinds of prey caught in ar- tificial spider webs with considerations of how orb-webs trap prey. Bull. British Arachnol. Soc., 5:29-38. Christensen, A., R. Baum & P.N. Witt. 1962. Changes in spider webs brought about by mes- caline, psilocybin, and an increase in body weight. J. Pharm. Exp. Therap., 136:31-37. Coddington, J.A. 1986a. The monophyletic origin of the orb web. Pp. 319-363, In Spiders: Webs, Behavior, and Evolution. (WA. Shear, ed.). Stan- ford Univ. Press, Stanford, California. Coddington, J.A. 1986b. Orb webs in “non-orb weaving” ogre faced spiders (Araneae: Dinopi- dae): a question of genealogy. J. Cladistics, 2: 53-67. Coddington, J.A. 1990b. Ontogeny and homology in the male palpus of orb weaving spiders and their potential outgroups, with comments on phy- togeny (Araneoclada, Araneoidea, Deinopoidea). Smith. Contrib. ZooL, 496:1-52. Coddington, J.A. & H.W Levi. 1991. Systematics and evolution of spiders (Araneae). Ann. Rev. Ecol. Syst., 22:565-592. Craig, C.L. 1987a. The significance of spider size to the diversification of spider-web architectures and spider reproductive modes. American Nat., 129:47-68. Craig, C.L, 1987b. The ecological and evolution- ary interdependence between web architecture and web silk spun by orb web weaving spiders. Biol. J. Linn. Soc., 30:135-162. Craig, C.L. 1988. Insect perception of spider orb webs in three light habitats. Fund, Ecol. 2:277- 282. Craig, C.L. 1990. Effects of background pattern on insect perception of webs spun by orb weaving spiders. Anim. Behav., 39:135-144. Craig, C.L. & G.D. Bernard. 1990. Insect attraction to ultraviolet-reflecting spider webs and web dec- orations. Ecology, 71:616-624. Craig, C.L., B. Bernard & J. Coddington. 1994. Evolutionary shifts in the spectra properties of spider silks. Evolution, 48:287-296. Craig, C.L. & K. Ebert. 1994. Color and pattern in predator-prey interactions: the bright body colors and patterns of tropical orb-weaving spiders at- tract flower-seeking prey. Funct. Ecol., 8:616- 620. Craig, C.L. & C.R. Freeman. 1991. Effects of predator visibility on prey encounter. A case study on aerial web weaving spiders. Behav, Ecol. SociobioL, 29:249-254. Eberhard, WG. 1972. The web of Uloborus div- ersus (Araneae: Uloboridae). J. ZooL Soc. Lon- don, 166:417-465. Eberhard, W.G. 1986. Effects of orb- web geometry on prey interception and retention. Pp. 70-100, In Spiders: Webs, Behavior, and Evolution. (WA. Shear, ed.). Stanford Univ. Press, Stanford, California. Eberhard, W.G. 1988. Combing and sticky silk at- tachment behavior by cribellate spiders and its taxonomic implications. Bull. British Arachnol. Soc., 7:247-251. Eberhard, W.G. 1989. Effects of orb- web orienta- tion and spider size on prey retention. Bull. Brit- ish Arachnol. Soc., 8:45-48. Eberhard, WG. 1990. Function and phytogeny of spider webs. Ann. Rev. Ecol. Syst., 21:341-372. Eberhard, WG. & F. Pereira. 1993. Ultrastructure of cribellate silk of nine species in eight families and possible taxonomic implications. (Araneae: Amaurobiidae, Deinopidae, Desidae, Dictynidae, Filistatidae, Hypochilidae, Stiphidiidae, Tengel- lidae). J. Arachnol., 21:161-174. Foelix, R.F. 1996. Biology of Spiders. Oxford Univ. Press, New York. Harvey, PH. & M.D. Pagel. 1991. The Compara- tive Method in Evolutionary Biology. Oxford Univ. Press, New York. Huey, R.B. & A.F. Bennett. 1986. A comparative approach to field and laboratory studies in evo- lutionary biology. Pp. 82-96, In Predator-prey Relationships: Perspectives and Approaches for the Study of Lower Vertebrates. (M.E. Feder & G.V. Lauder, eds.). Univ. of Chicago Press, Chi- cago. Kohler, T. & F. Vollrath. 1995. Thread biomechan- ics in the orb-weaving spiders Araneus diade- matus (Araneae, Araneidae) and Uloborus wal- ckenaerius (Araneae, Uloboridae). J. Exp. ZooL, 271:1-17. Levi, H.W. 1985. The spiny orb- weaver genera Mb 306 THE JOURNAL OF ARACHNOLOGY crathena and (Araneae: Araneidae). Bull, Mus. Comp. ZooL, 150:429-618. Lin, L.H., D.T, Edmonds & E Vollrath. 1995. Structural engineering of an orb-spider’s web. Nature, 373:146-148. Lubin, Y.D. 1986. Web building and prey capture in the Uloboridae. Pp. 132-171, In Spiders: Webs, Behavior, and Evolution. (W.A. Shear,ed.). Stanford Univ Press, Stanford, California. Maddison, W.P. & D.R. Maddison. 1992. Mac- Clade: Analysis of Phylogeny and Character Evolution, Version 3. Sinauer Associates, Inc., Sutherland, Massachusetts. Opell, B.D. 1979. Revision of the genera and trop- ical American species of the spider family Ulo- boridae. Bull. Mus. Comp. ZooL, 148:443-549. Opell, B.D. 1990. The material investment and prey capture potential of reduced spider webs. Behav. EcoL SociobioL, 26:375-381. Opell, B.D. 1993. What forces are responsible for the stickiness of spider cribellar threads? J, Exp. ZooL, 265:469-476. Opell, B.D. 1994a. Increased stickiness of prey capture threads accompanying web reduction in the spider family Uloboridae. Funct. EcoL, 8:85- 90. Opell, B.D. 1994b. Factors governing the sticki- ness of cribellar prey capture threads in the spi- der family Uloboridae. J. Morph., 221:111-119. Opell, B.D, 1994c. The ability of spider cribellar prey capture thread to hold insects with different surface features. Funct, EcoL, 8:145-150. Opell, B.D. 1994d. Factors affecting the diameters of axial fibers in cribellar threads of the spider family Uloboridae. J. ArachnoL, 22:12-18. Opell, B.D. 1995. Ontogenetic changes in cribel- lum spigot number and cribellar prey capture thread stickiness in the spider family Uloboridae. J. Morph., 224:47-56. Opell, B.D. 1996. Functional similarities of spider webs with diverse architectures. American Nat., 148:630-648. Peters, H.M. 1937. Studien am Netz der Kreuz- spinne (Epeira diademata CL), II: Uber die Hert- stellund des Rahmens, der Radialfaden, und der Hilfspiral. Zeits. Morph. Okol. Tiere, 33:128- 150. Peters, H.M. 1983. Stmktur und Herstellung der Fangfden cribellater Spinnen (Arachnida: Ara- neae). Verb, Naturweiss. Ver. Flamburg, 26:241- 253. Peters, H.M. 1984. The spinning apparatus of Ulo- boridae in relation to the structure, and construc- tion of capture threads (Arachnida, Araneida). Zoomorphology, 104:96-104. Peters, H.M. 1986. Fine structure and function of capture threads. Pp. 187-202, In Ecophysiology of Spiders. (W. Nentwig, ed,). Springer- Verlag, New York. Peters, H.M. 1992. On the spinning apparatus and structure of the capture threads of Deinopis sub- rufus (Araneae, Deinopidae). Zoomorphology, 112:27-37. Peters, H.M. 1995. Ultrastracture of orb-spiders' gluey capture threads. Naturwissen., 82:380- 382. Riecheit, S.E. & A.B. Cady. 1983. Patterns of re- source use and tests for competitive release in a spider community. Ecology, 64:899-913. Risch, P. 1977. Quantitative analysis of orb web patterns in four species of spiders. Behav. Genet., 7:199-238. Spiller, D.A. 1984. Competition between two spi- der species: an experimental field study. Evolu- tion, 65:909-919. Tillinghast, E.K., M.A. Townley, T.N. Wight, G. Uhlenbmck, & E. Janssen. 1993. The adhesive glycoprotein of the orb web of Argiope aurantia (Araneae, Araneidae). Materials Res. Soc., Symp. Proc., 292:9-23. Townley, M.A., D.T. Bernstein, K.S. Gallanger, & E.K. Tillinghast. 1991. Comparative study of orb-web hydroscopicity and adhesive spiral com- position in three areneid spiders. J. Exp. ZooL, 259:154-165. Vollrath, E 1992. Spider webs and silks. Sci. American, 266:70-76. Vollrath, F. & D.T. Edmonds. 1989. Modulation of the mechanical properties of spider silk by coat- ing with water. Nature, 340:305-307. Vollrath, E, WJ. Fairbrother, R.J.P. Williams, E.K. Tillinghast, D.T. Bernstein, K.S. Gallagher & M.A, Townley. 1990. Compounds in the drop- lets of the orb spider's viscid spiral. Nature, 345: 526-528. Vollrath, F. & E.K. Tillinghast. 1991. Glycoprotein glue beneath a spider web’s aqueous coat. Na- turwissen., 78:557-559. Wiehle, H. 1927. Beitrage zur Kenntnis des Rad- netzbaues der Epeiriden, Tetragnathiden und Uloboriden. Zeits. Morph. OkoL Tiere, 8:468- 537. Wise, D.H. & J.L. Barata. 1983. Prey of two syn- topic spiders with different web structures. J. Ar- achnoL, 11:271-281. Witt, P.N., C. Reed & D.B. PeakalL 1968. A Spi- der's Web. Springer- Verlag, New York. Manuscript received 19 March 1996, accepted 30 May 1997. 1997. The Journal of Arachnology 25:307-320 NATURAL HISTORY AND COPULATORY BEHAVIOR OF THE SPINY ORBWEAVING SPIDER MICRATHENA GRACILIS (ARANEAE, ARANEIDAE) Todd C. Bukowski and Terry E. Christenson: Department of Psychology, Tulane University, New Orleans, Louisiana 70118 USA ABSTRACT. We examine copulatory behavior and the reproductive natural history of the spiny orb- weaver, Micrathena gracilis. Censuses were conducted on free-ranging, individually marked spiders. After molting to adulthood, males induct sperm into their palps and then search for mates. Females inhabit solitary, individually-constructed webs. Males preferentially remain with penultimate-instar females, those about to molt and mate for the first time. After a female molts and constructs a viscid spiral, males build mating threads on which they court. After copulating, the male must dismount and reapproach the female to inseminate her second reproductive tract. Two copulations are therefore required for a complete mating between male and female. Some males, however, obtained only one copulation and two males often copulated with a given female. Staged encounters in the field revealed the important observation that when a male did copulate twice with a female, the duration of the second copulation was more than twice as long as the first. Shortly after the second copulation, the male inducted sperm into the palps and moved away. Females remained sexually receptive throughout their lives and apparently mated with any male. Females oviposited about 30 days after molting and mating. Egg sacs were cryptic in appearance and yet clutch mortality was high. Copulatory behavior is discussed in relation to this reproductive natural history. Spiders offer an intriguing model for the study of reproduction. To begin, the sexes of= ten differ dramatically in both morphology and behavior (Foelix 1980; Vollrath & Parker 1992). The female orb weaver, for example, is generally a relatively large, sedentary predator while the adult male is smaller and, at least as an adult, a wanderer. The shape and pres- ence and arrangement of sperm intake and fer- tilization ducts of the female spider’s repro- ductive tracts is thought to have a strong influence on male sexual maturation rates and the pattern of cohabitation with females. Con- duit spermathecae, those with separate insem- ination and fertilization ducts (entelegynes), are thought to promote a first male advantage in fertilization due to a serial ordering of sperm and a first-in/first-out usage pattern (Austad 1984). Consequently, males of such species usually mature before females and co- habit with penultimate-instar females ap- proaching the final molt and sexual maturity (Christenson & Goist 1979; Robinson & Rob- inson 1980; Jackson 1986; Watson 1990; Dodson & Beck 1993). Cul-de-sac spermathe- cae, those with common insemination and fer- tilization ducts (haplogynes), are thought to promote a last male advantage because the sperm from the last mating are nearer to the exit of the duct and a last-in/first-out usage pattern (Austad 1984). This is supported by Kastner & Jacobs (1997; but see Eberhard et al. 1993). In this case there should be no se- lective pressure for early male maturation and preferential cohabitation with females ap- proaching sexual maturity. Males mature at about the same time or after females and co- habitation with juvenile females is not noted (Huff & Coyle 1992; Eberhard et al. 1993) While the morphology of the female repro- ductive tract may influence male advantage patterns for fertilization of a female’s eggs, sexual maturation rates and male cohabitation patterns, the relationship between the female’s reproductive tract morphology and copulatory behavior is uncertain. It is known that males of species with cul-de-sac spermathecae often simultaneously insert both palps during cop- ulation (Foelix 1980). The patterns of palpal insertion and duration of copulation among species with conduit spermathecae show ex- treme variability within and between species (Robinson & Robinson 1980; Elgar 1995). Males can insert one palp several times before 307 308 THE JOURNAL OF ARACHNOLOGY switching to the opposite palp or insert each palp once. Perhaps features of the natural his- tory can influence, or at least help explain, male reproductive behaviors. Our work focuses on the natural history and copulatory behavior in the spiny orbweaving spider, Micrathena gracilis (Walckenaer 1805). The genus Micrathena Sundevall 1833 contains over 100 species distributed through- out the New World tropics, with three found in North America (Levi 1978, 1985). The two species native to Louisiana, M. gracilis and M. sagitatta (Walckenaer 1841), have one generation a year at our study site. Micrathena are characterized by prominent spines on the female abdomen. The male is only a fraction of the size of the female with little similarity in form. Female Micrathena are also entele- gynes, having separate insemination and fer- tilization ducts (Levi 1978, 1985). Limited ob- servations of mating in captive M. gracilis were recorded by Montgomery (1903). Ob- servations have also been conducted on M. gracilis macrohabitat preferences (Hodge 1987a), site tenacity (Hodge 1987b) and prey selection (Uetz & Biere 1980; Uetz & Hart- sock 1987). During the course of field observations of M. gracilis, we noted that a complete mating between a given male and female entails two copulations (insertions), one for each of the two male and female copulatory organs. After the first copulation, the male must dismount and reapproach the female to copulate again. Staged encounters revealed the important ob- servation that the durations of the two copu- lations are asymmetrical. As we report here, the duration of the second copulation is more than twice as long as that of the first. Others have noted such differences in copulatory du- rations (Bristowe 1929; Huber 1993, 1995; Sasaki & Iwahashi 1995); but to our knowl- edge, no one has yet addressed how a copu- latory pattern might be related to a species’ natural history. We conducted a census of marked, unre- strained animals and observed staged encoun- ters to describe copulatory behavior of M. gracilis and to place it within a framework of reproductive natural history. We describe the timing of sexual maturation, architecture of the female web, behavior of males on the fe- male web, courtship and copulation, the like- lihood that copulation will occur, sperm in- duction, female oviposition and egg sac mortality. We then discuss how male copula- tory pattern may relate to these natural history data. METHODS Study site. — -Observations were conducted at the F. Edward Hebert Center of Tulane Uni- versity, 20 km south of New Orleans, Loui- siana. The studies were conducted on a 30 X 40 m plot in a hardwood, bottomland forest M. gracilis occur there in relatively large numbers. The study area, located next to a la- goon, is frequently flooded during the spider’s mating season. Census procedures. — To describe repro- ductive natural history, census observations were made during June through September, 1990 and 1991. Census animals were individ- ually marked with fast-drying acrylic paint and were observed daily. Females were marked on the tips of the spines and males on the dorsal surface of the abdomen. When an- imals molted they were re-marked. Paint- marking had no obvious effects on the spi- ders’ behavior. All web sites were tagged. During the 1990 field season, a total of 143 females was observed between 0700-1600 h; no males were marked. During the 1991 field season, a total of 102 females and 105 males was observed between 0700-1600 h. Each day the census area was thoroughly scanned for animals and all animals were briefly ob- served at least once. All unmarked animals located in the census area were marked. Noted were the presence of a viscid spiral (prey catching surface) and web support strands, presence and identity of the female and males on a web, pattern of residency, molting, mat- ing, oviposition, and disappearance or move- ment from the web site. For analyses of male residency on a female’s web we used the term “male days”. For example, if two males were present on a particular day on a given female’s web, we scored this as two male days for that female on that day. Every other day in late July and August of 1990, two plots about one km apart were searched for new egg sacs. Tagged intact sacs were examined daily. Five sacs were collected mid-season and their clutches were removed and weighed on a Mettler analytical balance. Eggs were then separated from each clutch with a paintbrush and 10 groups of 20 eggs BUKOWSKI & CHRISTENSON— NATURAL HISTORY OF MICRATHENA GRACILIS 309 {n — 200 eggs per clutch, most of the eggs in the clutch) were weighed. The length and width of 10 eggs from each clutch were also measured. Timing of first sperm induction. — Five penultimate-instar males approaching the final molt were placed in separate 250 ml collection vials that contained two small twigs. The vials were examined daily for molting or exoskeL etons. The date molted was recorded. Three days later the males were brought into the lab and examined for sperm content. The methods for determining sperm content are reported elsewhere (Bukowski & Christenson 1997). Here we simply note whether sperm were present in the palps or not. Procedures for staged matings.— To fa- cilitate the observation of complete mating se- quences, we conducted staged encounters be- tween males and females in 1990 and 1991. To be certain females were virgin, they were monitored for an impending molt, which is preceded a day or two by failure to construct a viscid spiral. The abdomen of a penultimate- instar female approaching the final molt is spherical with relatively short spines while that of the newly-molted female is elongate with longer spines. Penultimate-instar females were placed in 250 ml plastic collection vials where they molted to adulthood. This ensured they did not copulate overnight when not ob- served. The females were released the next day at their original web site. Males were col- lected from the webs of penultimate-instar fe- males and usually held a couple of days prior to testing. After collection they were exam- ined under a dissecting microscope for bodily damage. Over the 1991 field season, 176 males were also examined for bodily damage to determine if they are typically injured in male-male interactions. The reproductive his- tories of the males were not known. Staged encounters were initiated only under dry conditions. Females were released at 0800 h and allowed to build webs. A randomly cho- sen male was placed on an upper frame thread after the female had constructed the viscid spi- ral. Copulation is defined as the interval be- tween palpal insertion and male dismount from the female, usually occurring immedi- ately after removal of the palp. The interco- pulatory interval refers to the amount of time between an animal’s first and second copula- tion. Frequency and duration of copulation: Males {n = 20) were each presented to a fe- male {n = 20) and allowed to copulate twice and depart the web. Males were observed until after sperm induction or until Wi h had elapsed. Both male and female were then col- lected and weighed (wet weight) that evening in the laboratory on a Mettler analytical bal- ance. Both males and females were released at the field station the following day. Phases of copulation: We examined in de- tail the phases (inflation state of the hemato- docha) of the two copulations. The males {n = 13) used were taken from studies involving other factors that influence frequency and du- ration of copulation (Bukowski & Christenson unpubl. data). We recorded how long the he- matadochae were inflated, the length of time deflating, intervals between deflation and palp removal, and palp removal and dismount. Prolonged second copulation: We tested whether the second copulation would be pro- longed when a male {n = 15) copulated with one palp with one virgin female and was then given another virgin female. Statistics. — All summary statistics are re- ported as X ±SD. RESULTS Sexual maturation. — Mating occurred from mid-June to mid- August. Early in June 1991 all females {n = 24) and all (n = 19) but one male in the study area were juveniles. By late June, only 10.7% of the females {n = 51) had matured compared to more than 70% of males in = 21, x" = 24.97, P < 0.00001). Sex ratio appeared to change over time as well. In early June, the ratio was nearly at unity in = 24$, n = 19 <3, 1.3:1; ^ 0.58, P = 0.44) and by late June females in = 51) outnumbered males in = 21; 2.4:1; x^ ^ 5.4, P = 0.02). Wandering adult males are difficult to find so these numbers represent, for the most part, males on females’ webs. After molting to adulthood, males remained on a single strand of their last web for 3.0 ±1.2 days in = 22). By this time, the dorsum remained white, but other body parts had turned from a gray to a rusty-red or black. In contrast, females were tan, black and white, or black and appeared not to change color at sexual maturation. Adult males weighed 3.3 ±0.3 mg in = 20) and newly-molted females 310 THE JOURNAL OF ARACHNOLOGY weighed 45.0 ±6.6 mg {n = 20). Adult males appeared not to feed, although they can drink water from leaves or silk surfaces. Timing of first sperm induction. — It is not known when or where free-moving males first induct sperm to the palps. Presumably this is done before they leave their last web site. Males presented to females did not induct sperm prior to mating, but did so after mating (see post-copulatory sperm induction below). However, the best evidence for sperm induc- tion prior to mate searching was found with males that had molted to adulthood in collec- tion vials. Three days after the final molt all males {n = 5) had sperm in both palps. The female’s web. — After sunrise, females built radially symmetric orbs usually within 4 m of the ground. The web is essentially two- dimensional with a relatively small (about 20 cm in diameter) viscid spiral situated within triangular frame threads and maintained under high tension (Uetz & Hartsock 1987). The vis- cid spiral is sloped between 0-45° off of ver- tical. There are no support or barrier strands adjacent to the hub. Females remained at the hub with the head down, abdomen tilted back, and the dorsum parallel to and facing the ground. This is an unusual position made pos- sible by the relatively long fourth femora (see photos and drawings in Levi 1985). At dusk, the female ingested virtually every strand ex- cept frame threads, on which she remained until morning. Two to four days before the final molt, the female removed her viscid spiral and did not construct another until after the molt. Gener- ally, all that remained was the top horizontal support strand and this was usually shortened within a day before the molt. Females would move within 2-3 cm of one end of the strand and there they molted. After shedding the exo- skeleton, the top foundation strand was lengthened and used as a foundation thread for the next viscid spiral. The exoskeleton was removed from the molting thread and recon- nected on the upper foundation strand near the edge of the viscid spiral. This pattern of molt- ing behavior was similar for juvenile females and males. Census females in the penultimate instar {n = 17) remained at their web sites for 11.4 ±5.8 days and moved 0.35 ±0.6 times be- tween the penultimate and final molts. The in- terval between the penultimate and final molts Table 1. — Census females: number of census fe- male observation days, number of female observa- tion days with one or more males present, and num- ber of observation days that males were found on the webs of ante-penultimate instar or younger fe- males, penultimate-instar females and adult fe- males. Some individual females are represented in more than one age category. Female instar Number of days with at Number least of female one male days present Number of male days < Ante-penultimate 621 30 38 Penultimate 726 217 297 Adult 174 7 8 was 15.4 ±2.1 days. Only 29% ever moved during this time. Adult census females {n = 81) remained at their web sites for 19.8 ±12.3 days and moved 2.2 ±1.3 times before death or disappearance. Male residency on females’ webs. — Of 73 individually-marked adult males, 40 were marked the day of their final molt or were found with an exoskeleton. Nearly half of the males {n = 19) that were marked the day of their final molt were never found after leaving their tagged sites. Most marked males were found on the web of one {n = 28) or two {n “ 15) females, but some were found on the webs of three {n = 8), four {n = 2) or seven (n = 1). Overall, males were observed to visit 1.3 ±1.25 females. Males were more likely found with penul- timate-instar than with younger juvenile or non- virgin adult females (Table 1; ~ 170.8, P < 0.00001). Over the census period, only 21 males were noted with ante-penultimate fe- males and they stayed 2.2 ±2.1 days. Seven of those males were present when the female molted to the penultimate instar and all left the web that day. Males {n = 80) did not re- main significantly longer (x = 3.3 ±2.6 days) with penultimate-instar than ante-penultimate instar females (Fj 99 — 2.98, P = 0.09). Males were infrequently found with adult females; they were always observed in copula. As each female approached her final molt the number of males on her web increased. We sampled three periods of equal length dur- ing the penultimate instar: (1) just after the BUKOWSKI & CHRISTENSON— NATURAL HISTORY OF MICRATHENA GRACILIS 311 molt to the penultimate instar, (2) mid-instar, and (3) just prior to the final molt. The dura- tion of each period was not constant across females. The durations of the periods were slightly different for some females and were based {post hoc) on the number of days a giv- en female did not construct a viscid spiral pri- or to the final molt, 3.6 ±1.3 days. Therefore, the average number of days for each of the three periods is also 3.6 ±1.3 days. Male pres- ence significantly differed across these three time periods (male days, F232 ^ 14.8, P < 0.0001; number of males, ^2,32 ^ 21.46, P < 0.0001). Post-hoc means comparisons showed that male presence increased with each time period. Few males were found with females that had just molted to the penultimate instar (0.2 ±0.7 male days, 0.1 ±0.3 males). A few more males were present at a point midway through the penultimate instar (1.5 ±2.3 male days, P “ 0.056; 0.6 ±0.9 males, P “ 0.032). Even more males were found with the female immediately prior to the final molt (3.6 ±2.9 male days, P ~ 0.0018; 1.5 ±1.1 males, P = 0.0002). Of 33 known penultimate-instar census fe- males, 79% {n = 26) had at least one male (x ^ 1.6, range 1-6) present the day before they molted. Males were usually stationary and found near the ends of foundation or periph- eral strands. One to two days prior to the final molt, when no viscid spiral was constructed, males seemed to become more active and to be located nearer the female. A detailed de- scription of intermale encounters was difficult to attain with the unaided eye, given male size and speed of engagement. Male encounters might be described as chases with brief phys- ical contact. Males generally maintained all eight legs and examination revealed no signs of bodily damage. Of the 176 adult males col- lected from the webs of penultimate-instar fe- males (1991) and examined under a dissecting microscope, only five males were missing a leg. Five others were missing a palp. Courtship and copulation. — Just after the female’s final molt, when she became sexually receptive, she mated while on a single strand or after she constructed a viscid spiral. In the latter case, the male constructed a mating thread between the end of the primary foun- dation strand and the outer end of a radial strand. The male courted vigorously by bouncing, bobbing and abdomen wagging, as Figures 1-3. — Lateral view of copulatory mount- ing of Micrathena gracilis. Male spider is shaded and female spider is white. 1. Female acceptance posture and male approach; 2. Insertion; 3. Final copulatory position. See text for details. defined for other Micrathena species by Rob- inson & Robinson (1980). The sequence of events leading to insertion and the copulatory position of M. gracilis is very similar to that of M. schreibersi (Perti 1833), also described by Robinson & Robinson (1980). When a female at the hub of a viscid spiral responded to a courting male, she moved across the viscid spiral, on the radial strand connected to the male’s mating thread. She moved onto the mating thread, and then let go with the first (I), second (II), and sometimes 312 THE JOURNAL OF ARACHNOLOGY third (III) pairs of legs (see Fig. 1). If the fe- male were hanging by legs II-IV, the male bit and pulled at legs II until the female released them from the mating thread. This acceptance posture, necessary for coupling, placed the fe- male at the proper angle for insertion, about 40-45° below the horizontal silk strand. The male then moved toward the female so they were head to head. Unsuccessful attempts at insertion were often followed by the male jumping from the female. The male then hung from his dragline, which was connected to the mating thread. When the male inserted one palp, the fe- male grasped the male’s abdomen with legs I and II and chelicerae and appeared to pull him in further toward her epigynum (see Fig. 2). The female then rubbed legs I and II against his ventrum. Through an apparently joint ef- fort, the male was flipped over 180° so that they were positioned ventrum to ventrum and facing essentially the same direction (see Fig. 3). The male was positioned with his cepha- lothorax midline at the epigynum and his ab- domen at an angle of 30-45° to the major axis of the female’s abdomen. A male that was po- sitioned over the female’s left side had the right palp inserted into the female’s right gen- ital pore. The male’s body was bent at the ped- icel, following the contour of the female’s body. The male was connected by a thread to the female’s abdomen which was later used when dismounting. Once inserted and flipped over onto the fe- male’s abdomen, the male was firmly locked in place. Shortly after being properly posi- tioned, the hematodocha expanded to full size. As with females of many Micrathena species, M. gracilis females have an epigynal protu- berance (scape) oriented just below and me- dial to the spermathecae (Levi 1985). During copulation, the male’s hematodocha nearly surrounded the end of this structure. Sclero- tized parts of the palp closely gripped the scape at defined indentations. No contractions or changes in the hematodocha could be seen with the unaided eye and neither males nor females showed rhythmic movements while in copula. After copulation was initiated, the female returned to the hub. The female’s mobility was not compromised and she was able to capture and feed on prey items while mating. Within a few minutes the hematodocha deflated and within a few seconds the male removed the palp from the female’s copulatory pore. Oc- casionally a male would begin pulling at the engaged palp at about the time most males would dismount. If the palp were not removed immediately, the male usually remained in- serted for a relatively long period, apparently stuck. On a few rare occasions free-moving females in the field were noted with a dis- membered palp in the epigynum. Complete insemination of the female re- quires two distinct copulations. Due to the specific orientation required for insertion, males must dismount the female in order to re-mount and copulate a second time. Thus, once mounted, males intromit and copulate once with one palp and can inseminate only one spermatheca during that mount. If more than one male were present with a receptive female, they would sometimes sever one another’s mating threads, occasionally causing a male to fall from the web (Bu- kowski & Christenson unpubl. data). If one was engaged in copulation, another would of- ten court, causing the female to move out onto his mating thread and exhibit the acceptance posture. This male would approach and at- tempt insertion which was precluded by the position of the copulating male. It then bit at the legs and palps of the engaged male but never dislodged it. A few males severed web foundation strands, causing the web to col- lapse. Method of dismounting the female. —At the termination of copulation, the male {n — 19) either climbed {n = 10) off the female and moved up across the viscid spiral to the top frame thread or it jumped {n = 9) off and hung below the web, connected to her abdo- men by the dismount thread. Movement of males opting for the former tactic elicited vig- orous jerking and pursuit by the female. After reaching a foundation strand, the male would immediately construct a mating thread and court. During staged encounters, the interco- pulatory interval for males that climbed was 203 ±241 sec (n = 9). Males that jumped off occasionally attempt- ed to climb back up the dismounting strand. The female usually jerked her abdomen vig- orously, breaking the dismount strand and dis- connecting the pair. Those that remained sus- pended released silk that usually connected to the bottom foundation strand or to nearby veg- BUKOWSKI & CHRISTENSON— NATURAL HISTORY OF MICRATHENA GRACILIS 313 etation. If the released silk did not connect to the web, the male sent out additional strands until one attached. If several attempts failed, the male moved away from the web. Once the released silk connected, the male climbed onto the foundation strand, constructed a mating thread, and courted. During staged encounters, the intercopulatory interval for males that jumped (440 ±278 sec, n — 6) was not sig- nificantly shorter than for males that climbed (203 ±241 sec, w = 9, - 3.1, P - 0.10). However, one male that climbed off the fe- male courted the female nearly immediately, but was not able to insert until 834 sec later. When this outlier is removed, the intercopu- latory interval for males that climbed (123.4 ±49 sec, w = 5) was significantly shorter than for males that jumped (Fij2 = 10.23, P = 0.008), Males did not groom legs or palps be- tween copulations. The likelihood of obtain- ing a second copulation did not differ for males that climbed or jumped (x^ ^ 0.72, P = 0.40). Frequency of re-mating.— -Females were sexually receptive throughout their lives and mated with virtually any male that encoun- tered her web. Of the 57 adult census females (1990) that were observed briefly each day until oviposition, 34 (60%) were observed to mate on the day of the final molt, generally within 1 h after the molt. An additional 23 were observed to re-mate at a later date; 15 did so at least once, three twice, four thrice and one five times. Newly-molted females of- ten alternated copulations with two or more males. Overall, census females were observed to mate with 1.7 ±1.1 males. The frequency of female mating is likely to be much higher, however, because copulations are brief and males do not remain after mating. Copulations during staged encounters.— Staged encounters were held to determine more accurately the likelihood and duration of copulation under more controlled conditions. Frequency and duration of copulation: Of the 20 newly-molted females presented a male, 19 were receptive. Fifteen (78.9%) cop- ulated with the male on both sides or repro- ductive tracts, while four (21.1%) copulated on only one side. Of these four females, one did not respond to the male’s courtship after the first copulation, one had a male possessing only one palp (the male pulled the other palp off after it was examined under the micro- scope and before it copulated) and he did not court a second time; and the other two males jumped after the first copulation and became disconnected. The female that did not copulate was presented a total of four males. She sev- ered their mating threads or bit and threw the males to the ground. For the second copulation, males employed the unused palp and inserted it into the virgin epigynal opening. All males {n = 15) copu- lated for a longer duration during the second coupling (1448 ±1265 sec) than the first (630 ±178 sec, Fij4 = 5.93, P = 0.029). The cop- ulatory duration for males that mated on one side only (548 ±232 sec, n = A) did not sig- nificantly differ from the duration of the first copulation of males that mated on both sides (^1,17 = 0.56, P = 0.46). All males terminated the second copulation by jumping off the fe- male and hanging by the dismount strand. Af- ter mating, males departed the web, thus post- mate defense was not noted. Phases of copulation: We examined in de- tail the inflation phases of copulations of an additional group of males {n = 13). Once the palpal conductor was inserted, the hematodo- chae inflated, reaching full size (several times larger than the unused palp) by the end of the first minute. At this time the hematodochae appeared a translucent tan color. After several minutes (see Table 2 for time course of cop- ulatory events), the hematodocha turned opaque white as it began deflating. The con- ductor remained inserted for a few nfinutes after deflation. It was then removed and the male dismounted. Overall, the second copu- lation was significantly longer than the first (Fi,i2 = 30.17, P < 0.0001; Table 2). Signif- icantly different amounts of time were spent in each phase within a copulation (F336 ^ 29.99, P < 0.0001); most of the copulatory time was spent with the hematodochae inflat- ed and less time was spent in each successive phase. The copulation (first or second) by phase (the four phases) interaction was sig- nificant (F336 = 3.06, P = 0.041); while the duration of the inflation phase was twice as long in the second copulation, the time taken for deflation and the time from deflation to palp removal showed a five and eight-fold in- crease, respectively, in the second copulation (Table 2). Longer hematodochal inflation time did not necessarily result in a longer time to deflate. 314 THE JOURNAL OF ARACHNOLOGY Table 2. — Mean durations (seconds) of copulatory phases and mean copulatory durations for the first and second copulations with newly-molted virgin females (n = 13), First copulation Second copulation Mean SD Mean SD Duration of hematodochal inflation 507 127 1036 508 Duration of hematodochal deflation 95 86 408 454 End of deflation to palp removal 61 60 429 505 Palp removal to dismount 6 9 28 31 Total copulatory duration 670 201 1898 862 There were no significant relationships be- tween the duration of hematodochal inflation and the amount of time taken to deflate for the first (r — +0.29, P = 0.33) or the second cop- ulation (r = + 0.05, P = 0.88). There was no significant relationship between the durations of the first and the second copulations for males copulating twice (r = +0.09, P ~ 0.63, n = 28). However, one male’s palp became stuck in the female during the second copu- lation and he could not remove the palp until 5684 sec. When data on this male are re- moved, there was a significant positive rela- tionship between the duration of the first and second copulation (r = +0.44, P = 0.02; Fig. 4). We examined the distributions of the dura- tions of the first and second copulations for all animals that copulated twice with a single female (n = 28). We followed the method used by Suter (1990) to examine copulatory Duration of first copulation (sec.) Figure 4. — Relationship between the durations of the first and second copulations. Males (n = 27) that copulated longer for the first copulation also copulated longer during the second copulation (r = +0.44; Y = 248.8 + 1.91X). durations in a linyphiid. When the beginning of copulation is set at time 0, the times to completion of the first copulation fall along a straight line (r^ == 0.98; Fig. 5), This suggests that the copulations were nearly uniformly distributed. However, the second copulation was better described by a negative logarithmic (r^ = 0.95) than a linear function (r^ = 0.85). While measures of latency (in this case du- ration of copulation) are often positively skewed, first copulations (skew = 0.27; nor- mality = 0) were less positively skewed than second copulations (skew = 0.97). Prolonged second copulation: Males {n = 15) that had mated on only one side of a new- ly-molted female were presented to a second newly-molted virgin female. All males em- ployed the unused palp. They followed the same pattern of copulatory durations as a male mating on both sides of a single female. There Figure 5. —When the start of copulation is set at time 0, the number of males remaining in copula over time for the first copulation (o) is best de- scribed by a linear function (Y = 42.4 - 0.004A) and for the second copulation (•), a negative ex- ponential function (T = 53.146 (10“°-®®®^)Y). BUKOWSKI & CHRISTENSON— NATURAL HISTORY OF MICRATHENA GRACILIS 315 was no significant overall difference in the copulatory durations of males given one or two females (Fj 28 “ 0.29, P = 0.60). Overall, second copulations were significantly longer than first copulations (Fi 28 = 26.28, P < 0.0001). However, the durations of the first and second copulations did not differ as a function of the number of females with which the males mated (Fj 28 “ 0.75, P ~ 0.39). For all males that mated with two females, first copulations (572 ±147 sec) were much short- er than second copulations (1721 ±814 sec). Post-copulatory sperm induction. — ^After a male mated twice with a virgin female, it remained near, and rarely on, the female’s web for a short time, grooming palps, legs and gonopore. The male moved 1-3 m from the female’s web and resumed grooming until sperm web construction was initiated, 47.5 ±16.9 min {n = 15) after the final dismount. We do not know if males that mated only once with a female induct sperm prior to another mating. The sperm web was constructed on a single horizontal strand 5-60 cm in length. After making several passes along this line, the male laid another strand, 3~4 cm in length, basically parallel to and near the middle of the first. The two strands were held apart by legs I and II and legs III and IV. The latter pairs were extended and held out in front of the body so that an elongated, horizontal hexagon was formed. Silk was then laid in a zigzag manner between the two lateral sides of the hexagon. The area proximal to the cephalo- thorax was completely covered with silk, and it was to the ventral side of this area (approx- imately 1.0 X 1.5 mm) that the male applied the gonopore. After many (about 100) appli- cations, the male tipped its body up so it was perpendicular to the plane of the sperm web. He then reached over the dorsal side of the sperm web and applied the palps. It should be noted that one male (not included in this data set) had a broken fourth femur and could not keep the sperm web from collapsing before induction even though it constructed five webs. Palps were individually dipped a mean of 4.6 ±1.9 times before they were switched. Dips were extremely shallow, a travel distance of less than 1 mm. Observation of this behav- ior was difficult and made worse by the slightest breeze. Males (n = 15) exhibited 7.4 ±1.9 and 7.9 ±1.8 induction bouts for the right and left palp, respectively. Sperm induc- tion took 5.3 ±2.0 min (n ~ 15) to complete. The male then moved off the sperm web, which immediately collapsed, and departed. Observations were made only until the first sperm induction was completed, so males could have built additional sperm webs later. Oviposition.-— Females oviposited between July and October with most clutches laid in August and September. Of 101 marked adult females in 1990, 57 survived and remained in the study area to lay at least one clutch. Of the others, 13 were seen only the day they were marked and 31 disappeared before ovi- position, after an average of 17.8 ±9.6 days. Females {n = 8) that were observed molting, mating and ovipositing laid their first clutch an average of 30.7 ±8.2 days after the molt. Most oviposition occurred within 5 m of the web site and within 1-3 m of the ground (2.5 ±0.88 m, n = 97). Most egg sacs (66%, n = 63) were found in boxelder {Acer negundo), one of the more common trees in the area. The eggs were laid near the center of the underside of a leaf that was folded transversally and sealed tightly. The resulting sac was triangular in appearance with silk threads connecting the egg sac to the intersection of the leaf petiole and connecting branch. We have not observed M. gracilis constructing egg sacs but females left the web site sometime after dusk, and the procedure was usually completed by 0900 h the following morning. By this time most fe- males had returned to their previous web site. Females that had oviposited at least once {n = 57) produced 1.6 ±0.8 clutches before they died or disappeared. Examination of five clutches laid during the middle of the egg- laying season revealed 266.8 ±11.5 eggs. Clutches weighed 67.1 ±4.2 mg, and individ- ual eggs 0.25 ±0.02 mg. The eggs were ovoid, 0.80 ±0.03 mm {n = 50) in length and 0.66 ±0.02 mm in width. By multiplying the mean number of egg clutches (1.6) by the mean number of eggs per clutch (266.8), the average female might lay 425 eggs throughout her lifetime. M. gracilis egg sacs appeared to us to be cryptic, yet most suffered predation. Of 75 egg sacs found in the egg sac census areas (sacs laid by unmarked females), 62 (83%) were already destroyed at initial examination. Intact egg sacs (n = 13) lasted for a mean of 316 THE JOURNAL OF ARACHNOLOGY only 3.9 days before we found that they had been opened and contents removed. Another 22 clutches laid by individually-marked cen- sus females (the female was found on the egg sac) did not fare much better. Overall, these lasted a mean of 10 days, however, many (41%; «= 10) were eaten within 48 h of being laid. Four egg sacs lasted between 18-57 days before either being eaten or falling to the ground. Only two egg sacs were intact as of mid-October. They hatched 37 and 41 days, respectively, after being laid. Within two weeks of hatching the spiderlings had molted. When there was evidence of predation, the entire egg mass had usually been pulled from the sac. The white silk that normally sur- rounded the eggs was outside the sac, in an elongated cotton-like mass. Except for egg sacs that fell to the ground {n = 3), they all {n = 94) appeared to have been destroyed by a similar means of attack. DISCUSSION In many spider species, males mature in fewer molts than females. Consequently, males mature before females and are often smaller (Vollrath & Parker 1992). We found that these sex differences apply to M. gracilis. Sexual bimaturation is thought to be related to male sperm priority patterns (Parker 1984; but see Head 1995). When the first male to mate with a given female fertilizes most of her eggs, selection should favor males that mature early in the season so they are present when the female molts to adulthood and becomes sexually receptive (Austad 1984). Male ad- vantage pattern for fertilizing a female’s eggs has been determined for six entelegyne spi- ders, those with a heavily sclerotized female reproductive tract and separate sperm uptake and fertilization ducts. Most show basically a first male advantage (Jackson 1980; Vollrath 1980; Austad 1982; Martyniuk & Jaenike 1982; Christenson & Cohn 1988; Watson 1991; Masumoto 1993; but see Andrade 1996). Given that Micrathena is an entelegyne and shows early male maturation, we suspect it, too, will show a first male advantage pat- tern for fertilizing a female’s eggs. Male M. gracilis are found more frequently on the webs of penultimate-instar females, particularly those approaching the final molt. Jackson (1986), Watson (1990), Dodson & Beck (1993) and others (cited in Christenson 1984) have noted cohabitation with females just prior to their final molt and initial sexual receptivity. If M. gracilis does show a first male advantage pattern for fertilizing a fe- male’s eggs, the tendency of males to remain with penultimate-instar females is understand- able as the female mates for the first time just after her final molt. We suspect that male M. gracilis can monitor an impending molt. As has been noted by Dodson & Beck (1993), the monitoring of the female must be frequent or continuous so that the male does not miss the female’s molt and lose her to another male. We don’t know how a male recognizes such a female, but there are at least four possibili- ties. First, females do not build a viscid spiral within a few days prior to the final molt, so a changing vibratory environment could cue the male. This would not, however, help the male determine when molting has occurred because the viscid spiral is not constructed three days prior to the final molt. Second, M, gracilis have stridulatory organs (Hinton & Wilson 1970; Uetz & Stratton 1982) that could be used to signal an approaching molt. Such communication may be relatively cryptic to the human observer and we may have missed it. Third, Hill & Christenson (1988) note that smaller juvenile Nephila clavipes (Linne 1767) females are more aggressive toward the male than are penultimate-instar females, so males may be responding to female aggres- siveness. We did not note such a change in aggressiveness in M. gracilis females. Fourth, the female could produce a male-attracting pheromone. Female spiders are known to pro- duce pheromones (Olive 1982; Watson 1986; Prenter et al. 1994) that have only recently been isolated and identified (Schulz & Toft 1993). Male M. gracilis are unable exclude other males from approaching the female. At best, a male might interfere with another’s court- ship activities. It is worth noting that a male on a female’s web is cohabiting and not ac- tually defending or guarding the female. We think that there are two major reasons that guarding may not have evolved. First, guard- ing may not occur because it is impossible to guard the female and her web effectively. The viscid spiral of the female M. gracilis web is not held in place by a complex of barrier strands and is essentially two-dimensional. Thus, there is no central location near the fe- BUKOWSKI & CHRISTENSON— NATURAL HISTORY OF MICRATHENA GRACILIS 317 male a male could occupy and prevent access by other males. This hypothesis is further sup- ported by that observation that a male must attract a female to his location to mate via the mating thread. Competing males could entice the female to mate from any point along the perimeter of the viscid spiral. A male on one side of the viscid spiral is unable to prevent another male from accessing the female on the opposite side of the viscid spiral. In contrast, some orb-weavers, such as Nephila clavipes, construct an enduring three dimensional orb- web, one containing the viscid spiral held in place by many barrier or support strands. Many of the barrier strands connect near the hub opposite the female and a male can re- main there, centrally located, and fend off oth- er males as they approach from virtually any direction (Christenson & Goist 1979). Copu- lation occurs just after the final molt at the hub. Therefore, a male occupying the hub po- sition almost always mates while males at the periphery of the web rarely ever mate with that female (Christenson & Goist 1979). Fur- ther evidence of a lack of functional guarding is that the male cannot fend off rival males while copulating and he must dismount the female after the first copulation. The female responds to courting males, even when in cop- ula, so if courting is occurring during a dis- mount, the female can mate with another male. A second major reason that guarding may not have evolved is that guarding one mate might be less profitable than searching for and mating with other females and thus it has not been selected for in males. The interval be- tween mating just after the final molt and ovi- position is about one month. This is a rela- tively long period of time for a male to remain and defend a sexually active female against other male visitors. If there is a strong first male advantage pattern for fertilizing a fe- male’s eggs, then little might be gained from guarding a mate given that subsequent males will fertilize very few of the female’s eggs. In addition, the male’s investment with a partic- ular female might well be lost due to preda- tory pressure on egg sacs. Our data indicate that female M. gracilis lay one or two egg sacs that are likely not to produce spiderlings in the spring. Loss of investment through spi- derling mortality is a relatively understudied phenomenon. We agree with Pitnick & Mar- kow (1994) that in species where the female produces relatively few egg sacs that suffer relatively high mortality, a male would benefit from mating with as many females as possible rather than guard a single female. This would increase the likelihood that he would sire or contribute to a surviving clutch. Like many spiders, a M. gracilis male must mount the female twice to inseminate both of the female’s pores. Proximately, males must dismount the female between copulations be- cause of the relatively complicated process of genitalic coupling. Male spiders must often assume very specific orientations to the fe- male in order to insert the palp (Foelix 1980). The evolution of male genitalia is thought to be influenced by sexual selection on females (Eberhard 1985). Ultimately, females might influence the evolution of male specific inser- tion orientations and structures that prevent the male from inserting both palps in a single mounting. Coyle & O’Sheilds (1990) have suggested that female spiders might have evolved multiple spermathecae to prevent a male from monopolizing access to all of her sperm storage sites. We agree and suggest that female spiders might also have evolved spe- cific insertion orientation requirements to pre- vent a male from monopolizing both tracts in a single mounting. Such an arrangement might allow a female to gain information about the quality of a male during the first copulation and between the first and second copulation. She could then allow or refuse the male access to the opposite tract based on that information. A precondition for the kind of asymmetry in copulatory duration that we document here is a male mating once on each side of the female. Insertions separated by a dismount and preceded by overt courtship appears to be the rule in the spiny orbweavers (sub-family Gasteracanthinae; Robinson & Robinson 1980) and other Araneidae (Bristowe 1929; Robinson & Robinson 1980). What aspects of araneid natural history would drive the evo- lution of copulations consisting of one inser- tion on each side of the female? Our descrip- tive work indicates that general aspects of M. gracilis reproduction are fairly typical for the family, that is, females are solitary, sedentary predators while males move from female to female. Males do not remain and defend mates, they often copulate with several fe- males, and females mate with more than one 318 THE JOURNAL OF ARACHNOLOGY male. We suspect that their pattern of mating behavior, two copulations separated by a dis- mount with the second copulation being much longer than the first, is influenced by the ease with which a second male can copulate with the female within a relatively brief period of time. Under certain circumstances selection should favor males that rapidly transfer sperm to one reproductive tract in a single insertion (copulation) rather than over a series of inser- tions. If a male is required to dismount in or- der to re-insert in the same epiginal opening, a second male could usurp the first male on that side. Our observations suggest that the likelihood of such usurpations is related to particular elements of reproductive natural history: males mature before females and many males may simultaneously cohabit with a single female, a web structure that does not allow a male to defend the female from com- petitors, and the use of a mating thread to en- tice the female to mate. There is no central area on the web near the female that the male can occupy, consequently a male must entice the female to his location in order to insert. Male M. gracilis appear highly motivated to copulate with both tracts of the virgin fe- male. If the suspected first male advantage pattern in fertilization is related to the conduit shape of the spermathecae, then the sperm pri- ority pattern would be determined separately for each tract. The advantage would not go to the first male to mate with a female, per se, but to the first male to mate with a given tract. Therefore, a male should attempt to copulate on both sides. A male that copulates only once would leave the other virgin tract available for insemination by other males. Unless females can preferentially use sperm from one tract over the other at oviposition, the first male is likely to cede 50% or more of the fertiliza- tions to a second male that mates with the remaining virgin tract. The failure to obtain two copulations with a given female is likely to have important im- plications for the male’s subsequent reproduc- tive activities. About 25% of the males failed to copulate twice with a given female. These males would leave the females with one palp filled with sperm and one empty palp. A male in this circumstance might either refill the empty palp or mate with the next female using one full and one empty palp. It is unclear whether two copulations are required to trig- ger the onset of sperm induction and whether a male could preferentially fill the empty palp. Given the rhythmic and stereotyped organi- zation of sperm induction behaviors, it is un- likely that a male could preferentially fill one palp. The filling of the empty palp may, in turn, influence whether a male copulates once or twice with the next female. The distributions of the durations of first and second copulations suggest that different selective forces might be operating. The du- rations of the first copulations were short and nearly normally distributed which suggests that stabilizing selection is operating. The du- rations of the second copulation were longer and more positively skewed, suggesting direc- tional selection on longer copulatory dura- tions. Given that a male cannot exclude the advances of other males, selection should op- erate on speed and efficiency of sperm transfer during the first copulation so that he can switch sides before another male moves onto the web. That directional selection is operat- ing on longer second copulations suggests that the second copulation may serve somewhat different functions than the first. All phases of the second copulation were prolonged. We have shown that the prolonged copulation can facilitate sperm storage on both sides of the female and may serve a mate guarding func- tion as well (Bukowski & Christenson 1997, unpubl. data). Such differences in copulatory durations have been reported in other species of a number of spider families (Bristowe 1929; Huber 1993, 1995; Sasaki & Iwahashi 1995) and similar functions might be served. ACKNOWLEDGMENTS This paper was improved by comments by Petra Sierwald and two anonymous reviewers. An earlier draft of this paper was improved with comments by Paul J. Watson. Financial support was provided by the Robert E. Flow- erree Fund of Tulane University. Thanks to Robin Richards for support in and out of the field. LITERATURE CITED Andrade, M.C.B. 1996. Sexual selection for male sacrifice in the Australian redback spider. Sci- ence, 271:70-72. Austad, S.N. 1982. First male sperm priority in the bowl and doily spider, Frontinella pyramitela. Evolution, 36:777-785. BUKOWSKI & CHRISTENSON— NATURAL HISTORY OF MICRATHENA GRACILIS 319 Austad, S.N. 1984. Evolution of sperm priority patterns in spiders, Pp. 223-249, In Sperm Com- petition and the Evolution of Animal Mating Systems (R.L, Smith, ed.). New York: Academic Press. Bristowe, W.S. 1929. The mating habits of spiders, with special reference to the problems surround- ing sex dimorphism. Proc. ZooL Soc. London, 21:309-358. Bukowski, T.C. & T.E. Christenson. 1997. Deter- minates of sperm release and storage in the spiny orbweaving spider, Micrathena gracilis. Anim. Behav., 53:381-395. Christenson, TE. 1984. Alternative reproductive tactics in spiders. American ZooL, 24:321-332. Christenson, T.E. & J. Cohn. 1988. Male advantage for egg fertilization in the golden orbweaving spider, Nephila clavipes. J. Comp. Psych., 102: 312-318. Christenson, T.E. & K.C. Goist. 1979. Costs and benefits of male-male competition in the orb- weaving spider Nephila clavipes. Behav. Ecol. SociobioL, 5:87-92. Coyle, F.A. & T.C. O’ Shields. 1990. Courtship and mating behavior of Thelechoris karschi (Arane- ae, Dipluridae), an African funnelweb spider. J. Arachnol., 18:281-296. Dodson, G.N. & M.W. Beck. 1993. Precopulatory guarding of penultimate females by male crab spiders, Misumenoides formosipes. Anim. Be- hav., 46:951-959. Eberhard, W. 1985. Sexual selection and animal genitalia. Harvard Univ. Press, Cambridge, Mas- sachusetts. Eberhard, W., S. Guzman-Gomez & K. Catley. 1993. Correlation between spermathecal mor- phology and mating systems in spiders. Biol. J. Linn. Soc., 50:197-209. Elgar, M.A. 1995. The duration of copulation in spiders: Comparative patterns. Rec. Western Australia Mus., 52:1-11. Emlen, S.T. & L.W. Oring. 1977. Ecology, sexual selection, and the evolution of mating systems. Science, 197:215-223. Fincke, O.M. 1982. Lifetime mating success in a natural population of the damselfly, Enallagma hageni (Walsh) (Odonata: Coenagrionidae). Be- hav. Ecol. SociobioL, 18:405-412. Foelix, R.F. 1982. Biology of Spiders. Harvard Univ. Press, Cambridge, Massachusetts. Head, G. 1995. Selection on fecundity and varia- tion in the degree of sexual size dimorphism among spider species (class Araneae). Evolution, 49:776-781. Hill, E.M. & T.E. Christenson. 1988. Male resi- dency on juvenile female Nephila clavipes (Ara- neae, Araneidae) webs. J. Arachnol., 16:257- 259. Hinton, H.E. & R.S. Wilson. 1970. Stridulatory or- gans in spiny orb- weaving spiders. J. ZooL, Lon- don, 162:481-484. Hodge, M.A. 1987a. Macrohabitat selection by the orb weaving spider, Micrathena gracilis. Psyche, 94:347-361. Hodge, M.A. 1987b. Factors influencing web site residence time of the orb weaving spider Mi- crathena gracilis. Psyche, 94:363-371. Huber, B.A. 1993. Genital mechanics and sexual selection in the spider Nesticus cellulanus (Ara- neae: Nesticidae). Canadian J. ZooL, 71:2437- 2447. Huber, B.A. 1995. Genital morphology and copu- latory mechanics in Anyphaena accentuata (An- yphaenidae) and Clubiona pallidula (Clubioni- dae: Araneae). J. ZooL, London, 235:689-702. Huff, R.P. & F.A. Coyle. 1992. Systematics of Hy- pochilus sheari and Hypochilus coylei, two southern Appalachian lampshade spiders (Ara- neae, Hypochilidae). J. Arachnol., 20:40-46. Jackson, R.R. 1980. The mating strategy of Phi- dippus johnsoni (Araneae, Salticidae): II. Sperm competition and the function of copulation. J. Arachnol., 8:217-240. Jackson, R.R. 1986. Cohabitation of males and ju- venile females: A prevalent mating tactic of spi- ders. J. Nat. Hist., 20:1193-1210. Kastner, J.L. & E.M. Jakob. 1997. Last male sperm priority in a haplogyne spider (Araneae: Pholci- dae): Correlations between female morphology and patterns of sperm usage. Ann. Entomol. Soc. America, 90:254-259. Levi, H. 1978. The American orb- weaver genera Colphepeira, Micrathena and Gasteracantha north of Mexico. Bull. Mus. Comp. ZooL, 148: 417-442. Levi, H. 1985. The spiny orb-weaver genera Mi- crathena and Caetacis (Araneae: Araneidae). Bull. Mus. Comp. ZooL, 150:429-618. Martyniuk, J. & J. Jaenike. 1982. Multiple mating and sperm usage patterns in natural populations of Prolinyphia marginata (Araneae: Linyphi- idae). Ann. Entomol. Soc. America, 75:516-518. Masumoto, T. 1993. The effect of the copulatory plug in the funnel-web spider, Agelena limbata (Araneae: Agelenidae). J. Arachnol., 21:55-59. Montgomery, T.H. 1903. Studies on the habits of spiders, particularly those of the mating period. Proc. Acad. Nat. Sci., Philadelphia, 55:59-149. Olive, C.W. 1982. Sex pheromones in two orb- weaving spiders, (Araneae, Araneidae): An ex- perimental field study. J. Arachnol., 10:241-245. Parker, G.A. 1984. Sperm competition and the evo- lution of animal mating strategies. Pp. 1-60, In Sperm Competition and the Evolution of Animal Mating Systems (R.L. Smith, ed.). Academic Press, New York. Pitnick, S. & T.A. Markow. 1994. Male gametic strategies: Sperm size, testes size, and the alio- 320 THE JOURNAL OF ARACHNOLOGY cation of ejaculate among successive mates by the sperm-limited fly Drosophila pachea and its relatives. American Nat., 143:785-819. Prenter, J., R.W. Elwood & W.I. Montgomery. 1994. Assessments and decisions in Metellina segmentata (Araneae: Metidae): Evidence of a pheromone involved in mate guarding. Behav. Ecol. Sociobiol., 35:39-44. Robinson, M.H. & B. Robinson. 1980. Compara- tive studies of the courtship and mating behavior of tropical araneid spiders. Pacific Insects Mon- ogr., 26:1-218. Sasaki, T. & O. Iwahashi. 1995. Sexual cannibal- ism in an orb-weaving spider Argiope aemula. Anim. Behav., 49:1119-1121. Schulz, S., & S. Toft. 1993. Identification of a sex pheromone from a spider. Science, 260:1635- 1637. Suter, R.B. 1990. Courtship and assessment of vir- ginity by male bowl and doily spiders. Anim. Behav., 39:307-313. Uetz, G.W. & J.M. Biere. 1980. Prey of Micrath- ena gracilis (Walkenaer) (Araneae: Araneidae) in comparison with artificial webs and other trap- ping devices. Bull. British Arachnol. Soc., 5: 101-107. Uetz, G.W. & S.P. Hartsock. 1987. Prey selection in an orb-weaving spider: Micrathena gracilis (Araneae: Araneidae). Psyche, 94:103-116. Uetz, G.W. & G. Stratton. 1982. Acoustic com- munication and reproductive isolation in spiders, Pp. 123-159, In Spider Communication, Mech- anisms and Ecological Significance (P.N. Witt & J.S. Rovner, eds.). Princeton Univ. Press, Prince- ton, New Jersey. Vollrath, E 1980. Male body size and fitness in the web-building spider Nephila clavipes. Z. Tier., 53:61-78. Vollrath, E & G.A. Parker. 1992. Sexual dimor- phism and distorted sex ratios in spiders. Nature, 360:156-159. Watson, P.J. 1986. Transmission of a female sex pheromone thwarted by males in the spider Lin- yphiia litigiosa (Linyphiidae). Science, 233:219- 221. Watson, P.J. 1990. Female-enhanced male compe- tition determines the first mate and principal sire in the spider Linyphia litigiosa (Linyphiidae). Behav. Ecol. Sociobiol., 26:77-90. Watson, P.J. 1991. Multiple paternity and first mate sperm precedence in the sierra dome spider, Lin- yphia litigiosa Keyserling (Linyphiidae). Anim. Behav., 41:135-148. Manuscript received 27 February 1996, accepted 5 March 1997. 1997. The Journal of Arachnology 25:321-325 MATING BEHAVIOR OF PHYSOLIMNESIA AUSTRALIS (ACARI, LIMNESIIDAE), A NON-PARASITIC, ROTIFER-EATING WATER MITE FROM AUSTRALIA Heather C. Proctor^: Department of Biology, Queen’s University, Kingston, Ontario K7L 3N6 Canada ABSTRACT. The diversity of sperm transfer behavior shown by water mites (Acaii, Hydrachnidia) is among the highest in the Arthropoda. However, sperm transfer has been described in fewer than 10% of water mite genera, all of them being Holarctic or cosmopolitan taxa. Here I describe mating behavior in Physolimnesia australis (Halik 1940), the sole representative of an Australian genus. P. australis is unusual in having larvae that do not parasitize insects, and in including rotifers in its diet. The highly dimorphic P. australis male responds to female presence by taking up an “embrace” posture in which he orients his opisthosoma and legs III toward approaching females. The female is caught in the embrace and her legs IV are secured by the modified tips of the male’s legs IV. The male deposits a glutinous mass on the female’s back, which she grooms towards her genital opening after being released. This mode of transfer differs from members of the confamilial genus Limnesia Koch 1836 in which males and females do not pair. Chelicerates show the greatest diversity of sperm transfer modes in the Arthropoda. In some taxa, males transfer sperm directly with a penis (Opiliones) or with secondarily de- rived genitalia (e.g., palps in Araneae). In oth- er groups, males transfer sperm indirectly by depositing spermatophores on a substrate, and either encouraging females to move over the sperm packets (e.g., Scorpionida) or allowing females to discover and take up sperm on their own (e.g., many Pseudoscorpionida) (Proctor et al. 1995). Finally, the horseshoe crabs (Merostomata) have external fertilization of eggs (Ruppert & Barnes 1994). Within the Chelicerata, some taxa exhibit greater behav- ioral diversity than others. For example, all spiders pair whereas pseudoscorpions may or may not have close associations between the sexes. The greatest variety of sperm transfer behavior occurs among the mites (Subclass Acari), and within this group the most diverse behavior is shown by the water mites (Sub- order Prostigmata, Hydrachnidia). With the exception of external fertilization, all possible modes of sperm transfer occur in the Hy- drachnidia from direct transfer via venter-to- * Current address: AES, Griffith University, Nathan 4111, Queensland, Australia, venter copulation (e.g., Midea Bruzelius 1854, Eylais Latreille 1796) to complete dissociation in which the sexes never meet (e.g., Hydrod- roma Koch 1837). Despite this amazing range of behavior, water mites have been poorly studied and mating observations have been published for only 24 of the more than 340 genera of water mites (Proctor 1992a,b). These observations have been limited almost entirely to species from North America and Europe, and there are no descriptions of sperm transfer in a non-holarctic genus. Here I de- scribe mating behavior in a monotypic genus of Australian water mites together with casual observations of its life cycle and diet. METHODS Physolimnesia australis (Limnesiidae) is a small (^ 1 mm) water mite found in the lit- toral zone of standing and slowly running wa- ter in Queensland and New South Wales (M. Harvey pers. comm.; Proctor pers. obs.). This species shows a strong sexual dimorphism in which males have a ventrally concave opis- thosoma and flattened terminal segments of legs III and IV (Fig. 1). Females are morpho- logically similar to species in the confamilial genus Limnesia and were previously described 321 322 THE JOURNAL OF ARACHNOLOGY as a species in this genus (Limnesia trituber- culata Viets 1955). I collected and observed P. australis on two occasions in 1995: in March from Cedar Creek approximately 50 km south of Brisbane, Queensland; and in Oc- tober from a large pond on the University of Queensland campus, St. Lucia. All observa- tions were made using a dissecting micro- scope and took place at the Department of En- tomology, University of Queensland. I separated mites according to sex and stage (adult and deutonymph) and maintained these groups in large well plates (well diameter = 35 mm, depth = 19 mm). Dipteran larvae (Culicidae, Chironomidae) and cladocerans (Moinidae) collected from a small pond on the University of Queensland campus were in- cluded as potential prey for the mites. I made behavioral observations on mites that had been collected as adults as well as those raised from deutonymphs in the lab. Voucher speci- mens are deposited in the University of Queensland Insect Collection, Department of Entomology, St. Lucia, Australia, 4072. RESULTS Life-cycle and predation. — Most water mites have a complex life-cycle with three ac- tive stages: the six-legged larva parasitizes adult aquatic insects, and is followed by two eight-legged predatory stages, the deuto- nymph and the adult (Smith & Cook 1991). PhysoUmnesia australis is an exception to this rule in that its larvae forgo the parasitic phase. Adult females collected from the field readily laid small clutches consisting of 1-8 eggs on the sides and bottoms of the wells. The eggs were large relative to the female (mean =134 fxm, SD = 8 jam, n — A). 1 observed that P. australis larvae remain within the coating of the egg clutch and transform directly into pre- daceous deutonymphs (via the inactive pro- tonymph). The newly emerged PhysoUmnesia deuto- nymphs were very small (body length « 250 |xm) and were unable to handle the large cla- doceran and dipteran prey I provided. Never- theless, they increased in size and transformed into adult mites. This energetic mystery was solved when I observed deutonymphs captur- ing and eating large phoretic rotifers that had been inadvertently introduced along with their moinid cladoceran hosts. The rotifers were identified as Brachionus variabilis Hempel 1896, a cosmopolitan epizootic species of 200-380 fjim in length (Koste & Shiel 1987). Thus to a 250 pm deutonymph, a single rotifer would be a substantial meal. Adult male and female F. australis were observed eating B. variabilis, as well as feeding on cladocerans and dipteran larvae. Neither the deutonymphs nor the adults appeared to forcibly remove the rotifers from their moinid hosts; rather, the mites captured rotifers that had detached from the cladocerans and were swimming freely in the wells. Mating behavior.— When a male P. aus- tralis was placed in a well that held females, he initially stood on the substrate and groomed himself vigorously by moving legs III and IV back over his dorsum and around to his venter. After a bout of grooming, the male was very still relative to the females, which were constantly crawling and swim- ming close to the substrate. In P. australis, as in most limnesiids (pers. obs.), crawling lo- comotion is accomplished by the first three pairs of legs, with legs IV moving in a con- stant fanning motion over the mite’s back, pre- sumably aerating the dorsal integument for gas exchange purposes. When a female bumped into the male or passed near him he immediately took up the “embrace” posture (Fig. 1). In this position the male’s opistho- soma was tilted at approximately 30° to the substrate, the flattened tips of legs III were touching and pressed against the substrate (thereby forming the circular “embrace”), and legs IV were held rigidly and vertically. The male oriented his embrace towards any fe- males that passed behind him. He also orient- ed towards other passing males and occasion- ally to swimming cladocerans. While in this posture the male was often approached by a female that -by accident or intent -crawled up behind the male and placed her capitulum over the flattened tips of the male’s legs III (Fig. 2). The male responded by elevating his opisthosoma to about 50° to the substrate and attempting to capture the fanning tips of the female’s legs IV in the flattened, scoop-like tips of his own fourth legs (Fig. 2). It was unclear how this capturing was achieved; pos- sibly, the long apical seta at the tip of the fe- male’s leg IV is secured by the groove in the male’s tarsus. When the male had captured both of the female’s legs she typically began to struggle. PROCTOR— MATING BEHAVIOR OF PHYSOLIMNESIA 323 Figures 1-4.— Mating behavior of Physolimnesia australis. 1, Male in the “embrace” posture with female approaching from behind; 2, Female within male’s embrace, male has captured the tip of her left leg IV in the tip of his modified leg IV; 3, Male has captured both leg tips and the pair is swimming jerkily; 4, Female grooms sticky material deposited by male on her back towards her ventrally located genital aperture. 324 THE JOURNAL OF ARACHNOLOGY However, the male gripped the female in the region of her 2nd or 3rd coxal plates with the tips of his legs III (Fig. 3). The pair typically left the substrate at this point and swam about in a jerky fashion. I observed at least 20 pair- ings that reached this stage; however, all but three were terminated when the female es- caped from the male’s grip after a few seconds of swimming. For the pairs that continued swimming, which usually lasted less than one minute, the male rubbed the concave ventral surface of his opisthosoma on the female’s dorsum. The male’s genital opening is located just behind the coxal plates of legs IV, and the rubbing of his venter against the female’s back probably represents deposition of the ejacu- late. In two of the three complete matings ob- served, the male slid around backwards to- wards the end of the female’s opisthosoma just before the pair separated. After the female escaped or was released from the male’s grip, she perched on the substrate and vigorously groomed back over her dorsum and around towards her ventrally located genital opening (Fig. 4). In two of the three complete matings, I observed opaque white material on the fe- male’s dorsum after she separated from the male (Fig. 4). DISCUSSION The mating behavior of Physolimnesia aus- tralis is very different from that of species in the genus Limnesia, the only other limnesiid genus for which reproductive behavior is known. Limnesia species show no sexual di- morphism save in body size (female larger) and degree of fusion of genital plates. In Lim- nesia spp., physical or chemical contact be- tween males and females is not required for spermatophore production and transfer (Witte 1991; Proctor 1992a). Rather, males main- tained alone will deposit spermatophores on a substrate, and females that later encounter them will take up sperm if so inclined. Proctor (1992a) called this mode of sperm transfer “complete dissociation”, and contrasted it with three other modes: incomplete dissocia- tion (physical or chemical contact between the sexes required for spermatophore deposition but no pairing between the sexes); pairing, in- direct transfer (male courts a given female, spermatophores deposited on substrate); and copulation (male places sperm in female’s sperm-receiving structure). The transfer mode of Physolimnesia australis appears to fall be- tween the third and fourth categories, as the male places the sperm on the female (as in copulation), but she must move it to her gen- ital opening (as in pairing, indirect transfer). This suggests that the categories of sperm transfer outlined by Proctor (1992a) may be too rigid to easily encompass all transfer be- haviors. It is not clear what motivates the P. aus- tralis female to enter the embrace of the male’s legs. In the water mite Neumania pap- illator (Unionicolidae), the female orients to- wards the male’s courtship signals because they resemble vibrations caused by prey (Proctor 1991). It is possible that male Phy- solimnesia australis engage in similar “sen- sory trapping” (sensu Christy 1995), perhaps by producing chemicals that mimic the scent of prey animals. The water mite Family Limnesiidae con- tains 23 genera, five of which are composed of species that are strongly sexually dimorphic {Physolimnesia, Timmsilimnesia K.O. Viets 1984, Centrolimnesia Lundblad 1935, Ptero- limnesia Viets 1942 and Acantholimnesia Viets 1954) (Cook 1974, 1980, 1986, 1988; Viets 1984). No two genera share the same types of male modifications, suggesting that sperm transfer with close contact between the sexes has evolved repeatedly in this family from an ancestral non-paired state, as has oc- curred in many other families of water mites (Proctor 1991). Physolimnesia australis is an unusual water mite in other aspects of its biology. Whereas most Hydrachnidia have a parasitic larva that acts as both a feeding and a dispersal stage, the larva of P. australis transforms to a pred- atory deutonymph without parasitizing an in- sect host. Suppression of the parasitic phase has been recorded in 29 species scattered through distantly related families of water mites, including a few confamilials in the ge- nus Limnesia (Smith & Cook 1991; Smith in press; H. Proctor pers. obs.). Like many spe- cies with non-feeding larvae, P. australis has a small adult body size, small clutch size and relatively large eggs for its body size (Cook, Smith & Brooks 1989; Smith in press). Al- though the loss of larval parasitism has inde- pendently arisen many times, it does not seem conducive to cladogenesis, as such lineages consist of single species (or populations) PROCTOR— MATING BEHAVIOR OF PHYSOLIMNESIA 325 whose closest relatives retain parasitic larvae (Smith in press). Although one might expect that loss of dispersal via parasitic larvae would occur in lineages that inhabited per- manent water bodies, there is no apparent pat- tern in relation to habitat: lineages without lar- val parasitism occur in streams, temporary ponds, and both littoral and planktonic habi- tats within lakes (Smith in press). It is not clear how, or indeed if, water mites with non- feeding larvae mites disperse to new bodies of water. The final strange aspect of P. australis' bi- ology is the inclusion of rotifers in its diet. Confamilials in the genus Limnesia have been observed feeding on a variety of invertebrates (crustaceans, insects, other mites) and even vertebrate prey (fish eggs) (Proctor & Prit- chard 1989); but to my knowledge, this is only the second observation of arachnids feeding on rotifers. One other species, an undescribed oribatid mite in the genus Aquanothrus En- gelbrecht 1975 (Ameronothridae), is believed to feed on rotifers based on the presence of undigested trophi (rotifer mouthparts) in the mites’ guts (R.A. Norton pers. comm.). Note added in proof: (a) Adults and nymphs also prey on nematodes and oligo- chaetes. (b) It is also possible that the male deposits spermatophores on his own legs III prior to taking up the embrace posture; sperm could thereby be inserted in the female’s gen- ital opening during the “nuptial swim,” and the gelatinous substance on her dorsum may be residual spermatophore material. ACKNOWLEDGMENTS Many thanks to Dave Walter (Entomology, University of Queensland) who offered lab space, room and board, and chauffeur service. Mark Harvey (Western Australian Museum) kindly provided me with access to his volu- minous water mite library, and both he and Bruce Smith (Ithaca College, New York) sent me copies of their own unpublished manu- scripts. This research was supported by a Queen’s University ARC grant. LITERATURE CITED Christy, J.H. 1995. Mimicry, mate choice, and the sensory trap hypothesis. American Nat., 146: 171-181. Cook, D.R. 1974. Water mite genera and subgen- era. Mem, American Entomol. Inst., 21:1-860. Cook, D.R. 1980. Neotropical water mites. Mem. American Entomol. Inst., 31:1-645. Cook, D.R. 1986. Water mites from Australia. Mem. American Entomol. Inst., 40:1-568. Cook, D.R. 1988. Water mites from Chile. Mem. American Entomol. Inst., 42:1-356. Cook, W.J., B.P. Smith & R.J. Brooks. 1989. Al- location of reproductive effort in female Arren- urus spp. (Acari: Hydrachnidia; Arrenuridae). Oecologia, 79:184-188. Koste, W. & J. Shiel. 1987. Rotifera from Austra- lian inland waters. II. Epiphanidae and Brachion- idae (Rotifera: Monogononta). Invert. Taxon., 1: 949-1021. Proctor, H.C. 1991. The evolution of copulation in water mites: a comparative test for nonreversing characters. Evolution, 45:558-567. Proctor, H.C. 1992a. Mating and spermatophore morphology of water mites (Acari: Parasitengo- na). Zook J. Linn. Soc., 106:341-384. Proctor, H.C. 1992b. Sensory exploitation and the evolution of male mating behaviour: a cladistic test using water mites (Acari: Parasitengona). Anim. Behav., 44:745-752. Proctor, H.C., R.L. Baker & D.T Gwynne. 1995. Mating behaviour and spermatophore morphol- ogy: a comparative test of the female-choice hy- pothesis. Canadian J. ZooL, 73:2010-2020. Proctor, H. & G. Pritchard. 1989. Neglected pred- ators: water mites (Acari: Parasitengona: Hy- drachnellae) in freshwater communities. J, North American Benthol. Soc., 8:100-111. Ruppert, E.E. & R.D. Barnes. 1994. Invertebrate biology, 6th ed. Saunders College Publ., New York. 1102 pp. Smith, B.P. In press. Loss of larval parasitism in parasitengonine mites. Exp. AppL, AcaroL, 21: 00-00. Smith, I.M. & D.R. Cook. 1991. Water mites. Pp. 523-592, In Ecology and Classification of North American Freshwater Invertebrates (J.H. Thorp & A.P Covich, eds.). Academic Press, New York. Viets, K. O. 1984. Uber Wassermilben (Acari, Hy- drachnellae) aus Australien. Arch. Hydrobiol., 101:413-436. Witte, H. 1991. Indirect sperm transfer in prostig- matic mites from a phylogenetic viewpoint. Pp. 107-176, In The Acari: Reproduction, Develop- ment and Life-history Strategies. (R. Schuster & PW Murphy, eds.). Chapman & Hall, New York. Manuscript received 30 April 1996, accepted 5 March 1997. 1997. The Journal of Arachnology 25:326-332 ON THE ABUNDANCE AND PHENOLOGY OF PALPIGRADI (ARACHNIDA) FROM CENTRAL AMAZONIAN UPLAND FORESTS J. Adis’, U. Scheller^, J.W. de Morais^ B. Conde'’ and J.M.G. Rodrigues^: ’ Max^ Planck-Institute for Limnology, Tropical Ecology Working Group, Postfach 165, D- 24302 Plon, Germany. ^Haggeboholm, Haggesled, S-53194 Jarpas, Sweden. ^Instituto Nacional de Pesquisas da Amazonia (INPA), C.R 478, 69.011-970 Manaus/AM, Brazil. '’Musee de Zoologie de TUniversite et de la Ville de Nancy, 34, rue Sainte- Catherine, F-54000 Nancy, France. ABSTRACT. The 745 palpigrads (micro-whip scorpions) collected in a 12 month period in the soil (0- 7 cm depth) of a secondary upland forest (120.1 ± 50.8 ind./mVmonth) and of a primary upland forest (29.4 ± 20.2 ind./mVmonth) near Manaus all belong to the species Eukoenenia janetscheki Conde 1993. About 75% of all specimens inhabited the mineral subsoil (3.5-7 cm depth) where monthly catches were negatively correlated with temperature and moisture content of the soil. Females were almost twice as abundant as males. The lack of a distinct reproductive period and the presence of juveniles and adults (both sexes) throughout the year indicate a plurivoltine mode of life. No specimens were caught on or above the soil surface. Abundances of E. janetscheki are compared with those of the Schizomida (tartarids) and Thelyphonida (vinegaroons) from the same study sites. E. janetscheki also represented the palpigrads obtained from the soil of three other upland forest types in Central Amazonia (0-14 cm depth) and accounted for 0. 1-0.3% of the total arthropod fauna. Terrestrial arthropods of Central Amazoni- an forests have been investigated for several years (Adis & Schubart 1984; Adis 1997; Adis et al. 1997b,c) cooperatively between the National Institute for Amazonian Research (INPA) at Manaus/Brazil and the Tropical Ecology Working Group at the Max-Planck- Institute for Limnology in Plon/Germany (Projeto INPA/Max-Planck). Data on abun- dance and phenology of Palpigradi sampled during a 12 month period in 1982/83 in both a primary and a secondary upland forest are now available, as their time-consuming taxo- nomical evaluation has been completed (Con- de 1993, 1997). Our data represent the very first contribution on the phenology of a trop- ical palpigrad species: E. janetscheki Conde 1993. Voucher specimens have been deposited at the Systematic Entomology Collections of the Instituto Nacional de Pesquisas da Ama- zonia (INPA) in Manaus, Brazil, at the Mu- seum d’histoire naturelle in Geneve, Switzer- land and at The Field Museum, Chicago, USA. Results presented are compared with abun- dance data of E. janetscheki from three other upland forest types near Manaus, and which were sampled between 1985 and 1991. STUDY AREA AND METHODS Palpigrads were collected between 1981 and 1983 in the course of ecological studies on Central Amazonian arthropods from two previously investigated and fully described forest types, all within 30 km of Manaus: (1) primary upland forest at Reserva Florestal A. Ducke (= Reserva Ducke; 2°55^S, 59°59'W) on the Manaus-Itacoatiara highway (AM-010 at km 26; cf Penny & Arias 1982), (2) sec- ondary upland forest at Rio Taruma Mirim (3°2'S, 60°17'W), a tributary of the Rio Ne- gro, where the vegetation was previously cut but unbumed (Adis 1992). Both forests are subject to a rainy season (December-May: av- erage precipitation 1550 mm) and a “dry” season (June-November: average precipitation 550 mm, but each month has some rain events; cf. Ribeiro & Adis 1984). The yellow 326 ADIS ET AL.— PALPIGRADI OF CENTRAL AMAZON 327 latosoil of the primary and secondary upland forests supported a 2-3 cm thick humus layer, interspersed with fine roots, and a thin surface covering of leaf-litter. One ground photo- eclector (emergence trap) and one arboreal photo-eclector for trunk ascents (funnel trap) were installed in both forests from December 1981 to December 1982 (Adis & Schubart 1984). Distribution of palpigrads in the soil was studied between September 1982 and Au- gust 1983 (Morais 1985; Rodrigues 1986). Twelve soil samples were taken once a month from each forest type at random along a tran- sect with a split corer (= steel cylinder with lateral hinges; diameter 21 cm, length 33 cm) which was driven into the soil by a mallet. The combined area of 12 samples represented 0.42 m^. Each sample of 7 cm depth was then divided into two subsamples of 3.5 cm each. Animals were extracted from subsamples fol- lowing a modified method of Kempson (Adis 1987). The monthly collection data of palpi- grads from the two soil layers in relation to changing conditions of precipitation, temper- ature and humidity of the air near the forest floor as well as moisture content, temperature and pH of the soil were statistically evaluated with a linear correlation test (Cavalli-Sforza 1972) using the original field data (cf. Morais 1985; Rodrigues 1986) of the previous month. In addition, the presence of palpigrads in tree crowns of the primary upland forest was test- ed by fogging canopies with pyrethrum (with and without synergist) during the dry and rainy seasons (July 1977, August 1991, Feb- ruary & August 1992, July 1994; Adis et al. 1984, 1997a). Palpigradi sampled were clas- sified as juveniles, subadults and adults (males and females) according to Conde (1984a,b, 1993, 1997). RESULTS Palpigradi obtained from different upland forest types in the vicinity of Manaus were represented by only one species: Eukoenenia janetscheki Conde 1993. The body length of adult males (without flagellum) reached 1 mm. A total of 146 specimens was collected in the primary upland forest at Reserva Ducke and 599 specimens in the secondary upland forest at Rio Taruma Mirim. Out of these, 92% could be identified to their developmen- tal stages. E. janetscheki was only found in the soil and never caught on tree trunks or in the canopy. No specimens were captured in ground photo-eclectors. In the primary upland forest, palpigrads represented 0.2% and in the secondary upland forest 0.6% of the total ar- thropods extracted from soil samples within 12 months (Acari and Collembola omitted; cf. Morais 1985; Rodrigues 1986). Their abun- dance in 0-7 cm soil depth was comparable to that of the Schizomida (Fig. 1). Most spec- imens of E. janetscheki inhabited the mineral subsoil (Fig. 2: 3.5-7 cm) and a few (22- 26%) the organic layer (0-3.5 cm depth). An average of 120.1 ± 50.8 ind./mVmonth was recorded in the secondary upland forest and 29.4 ± 20.2 ind./mVmonth in the primary up- land forest (0-7 cm depth). More than 50% of the total catch in both forests was repre- sented by adults (Fig. 2). Sex ratio (adult males to females) in the secondary forest was 1:1.8 (81% of the total adults could be sexed). In the secondary upland forest, the monthly abundance of E. janetscheki in the mineral subsoil (3.5-7 cm depth) was negatively cor- related with soil temperature, i.e., catch num- bers (in particular of subadults) decreased with increasing temperatures (total catch: r ^ -0.7454, P < 0.01; adults: r - -0.5926, P < 0.05; subadults: r = -0.8614, F < 0.001; n — 12, respectively). In the primary upland forest our data indicated a negative correlation of adults with the soil moisture content in the mineral subsoil (r — —0.5670, P < 0.\, n = 12). The total catches of specimens obtained during the dry season and the rainy season were similar: 66% versus 34% in the primary upland forest and 61% versus 39% in the sec- ondary upland forest, respectively. However, there was no distinct reproductive period in the secondary forest (where E. janetscheki was more abundant) because juveniles as well as adults (both sexes) occurred throughout the year (Fig. 3). These results indicate a plurivol- tine mode of life. DISCUSSION ! Comparable data on the abundance and ver- tical distribution of the soil fauna in three dif- ferent upland forest types of Central Amazo- nia were given by Adis and collaborators (Adis et al. 1987a,b; 1989a,b; Ribeiro 1994). Arthropods were collected to a soil depth of 14 cm during rainy and dry seasons and ex- tracted with the Kempson method as de- 328 THE JOURNAL OF ARACHNOLOGY — Palpigradi - - Schiiomida — “ Thelyphonida Figure 1. — Distribution of Palpigradi, Schizomida and Thelyphonida in the soil. Samples taken monthly at 0-7 cm depth between September 1982 and August 1983 in two upland forests near Manaus. (Total catch = 100% in each forest type; n = total number of specimens). Total precipitation per month given between sampling dates (= at the end of each month in the primary upland forest and in the middle of each month in the secondary upland forest). The low rainfall observed in early 1983 was due to a strong El Nino event (see Adis & Latif 1996). ADIS ET AL.— PALPIGRADI OF CENTRAL AMAZON 329 Primarv upland forest 90 n H-U2 11 adults □immatures (12,0 3 ” -pw- %)/ \ £ 1 / A 1 (63,4%) / S \ W24,6%) IS 0 1 • o 30 - IS J J = juveniles S = subadults A = adults 0 - 0 - 3.5 3.5 - 7.0 Soil Depth (cm) Secondary upland forest N=541 0 adults □ immatures 0 - 3.5 3.5 - 7.0 Soil Depth (cm) J = juveniles S = subadults A = adults Figure 2.- — Distribution of Eukeonenia janetscheki in the soil according to soil depth, and percentage of all developmental stages in two upland forests near Manaus. (Total catch = 100% in each forest type). Samples taken monthly at 0=3.5 and 3.5=7 cm depths between September 1982 and August 1983. n = total number of specimens. scribed above. Between 75% and 92% of all arthropods were found to inhabit the top 7 cm when Acari and Collembola were included in the total catch numbers and 69%-"84% when they were omitted. Data on Palpigradi E. janetscheki; see Conde 1997) are now avail- able. One study was conducted in 1985/86 in a secondary upland forest on yellow latosoil at the INPA campus in Manaus, where the veg- etation was previously cut but unbumed (Adis et al. 1987a,b). Palpigrads represented 0.2- 0.3% of the total arthropods when Acari and Collembola are included (dry season: 50.448 ind./m^, rainy season: 63.850 ind./m^) and 0.9% when they are omitted from the total catch numbers (dry season: 11.934 ind./m^, rainy season: 17.886 ind./m^). In the mineral subsoil (7~14 cm depth), the abundance of palpigrads was higher during the dry season (62% of the total catch; 62.6 ind./m^) but low- er during the rainy season (44% of the total catch; 72.2 ind./m^) when compared to the top 7 cm. Another study was made in 1990/91 in a secondary upland forest on yellow latosoil, about 50 km north of Manaus, where the veg- etation was previously cut and burned (Ribei- ro 1994). Palpigrads represented 0. 1-0.2% of the total arthropods when Acari and Collem- bola are included (dry season: 33.915 ind./m^, rainy season: 19.696 ind./m^) and 0.4-0.6% when they are omitted from the total catch numbers (dry season: 7.180 ind./m^, rainy sea- son: 7.777 ind./m^). In the mineral subsoil (7- 14 cm depth), the abundance of palpigrads was lower during the dry season (33% of the total catch; 9.6 ind./m^) but higher during the rainy season (78% of the total catch; 33.7 ind./ m^) when compared to the top 7 cm. A third study was conducted in 1988 in a primary forest on white sand soil, about 45 km north of Manaus (Adis et al. 1989a,b). Pal- pigrads represented 0. 1-0.2% of the total ar- thropods when Acari and Collembola are in- cluded (dry season: 57.703 ind./m^, rainy season: 74.255 ind./m^) and 0.5“0.9% when they are omitted from the total catch numbers (dry season: 14.119 ind./m^, rainy season: 15.023 ind./m^). The abundance of palpigrads in the mineral subsoil (7-14 cm depth) was higher during the dry season (86% of the total catch; 57.8 ind./m^) and during the rainy sea- son (89% of the total catch; 120.3 ind./m^) when compared to the top 7 cm. To which depth E. janetscheki occurs in the soil of the Central Amazonian upland forests is unknown. First studies below 14 cm soil depth in a primary forest on yellow latosoil and on white sand soil 45 km north of Manaus showed that soil layers in 20-30 cm depth were dominated by social insects, in particular Isoptera. Palpigradi were not found below 10 cm soil depth, probably due to handsorting of the soil samples (Harada & Bandeira 1994a,b). In Costa Rica, palpigrads were found during the rainy season in 15-20 cm soil depth with abundances of 350 ind./m^ in a forest and 75 ind./m^ in a coffee plantation. They represented 1.4% (800 ind./m^) and 330 THE JOURNAL OF ARACHNOLOGY Primary upland forest Secondary upland forest 40 n Subadults I 20 - a. m SONDJFMAMJJA 100 (0 g 50 p 0) Q. 1 (n i H . M , n - Subadults a SONDJ FMAMJJA 40 w i20H o 0) Q. (/) Juveniles EL JIL S O N D ^ 1982 EL n n I I I i 1 I I J FMAMJJA > 1983 < 100 w E 50 H a. Ui Juveniles ■n r 1 n n n i 1 n n 1 1 1 S O N D 1982 ' — p- — — p — p 1 \ 1 J FMAMJJA > 1983 < Figure 3. — Temporal occurrence of developmental stages of Eukoenenia janetscheki in the soil (N/m^ in 0-7 cm depth) in two upland forests types near Manaus. Monthly samples taken between September 1982 and August 1983. 0.3% (188 ind./m^) of the total arthropods col- lected to a depth of 20 cm, respectively (Ser- afino & Merino 1978). In Israel, Eukoenenia mirabilis (Grassi & Calandruccio 1885) was sampled to a soil depth of 15 cm in pine and oak forests. The 71 ind./m^ represented 0.07% of the total microarthropods obtained (Broza et al. 1993). In Californian pine forests paL pigrads were detected to a depth of 1.18 m (Price 1975). The absence of E. janetscheki in samples from ground photo-eclectors in Central Ama- zonia also indicates that the species is not ac- tive on the soil surface. This conclusion is supported by another study in the primary up- land forest at Reserva Ducke, in which no specimens were collected in 20 baited pitfall traps and in three or more ground photo-eclec- tors during a sample period of 12 month (Pen- ny & Arias 1982). Palpigrads are considered to be hygrophil- ous, photophobic, euedaphic inhabitants of soils or troglobites (Conde 1986, 1996; Janet- schek 1957; Mahnert & Janetschek 1970; Gruner 1993). E. janetscheki was not found in the soil of man-made pastures (0-14 cm depth) adjacent to upland forests in Central Amazonia. One reason might be the low hu- ADIS ET AL.— PALPIGRADI OF CENTRAL AMAZON 331 midity and high temperature of the soil around noon, particularly during the dry season (Adis & Franklin, unpubl. data). The abundance of E. janetscheki in the mineral subsoil of the primary upland forest at Reserva Ducke and of the secondary upland forest at Rio Taruma Mirim was influenced by both abiotic factors. Seraflno & Merino (1978) found no palpi- grads in the soil of a pasture in Costa Rica as well. E. janetscheki was also not found in Central Amazonian inundation forests (Adis & Schu- bart 1984, Adis & Ribeiro 1989). The pres- ence of non-winged terricolous arthropods in this biotope requires flood resistance, horizon- tal migration according to the high-water line or vertical migration onto the trunk or into the canopy in response to annual flooding of 5-1 months duration. Reproduction cycle and du- ration of life stages have to be synchronized with the periodic fluctuations in water-level (Adis 1997; Adis et al. 1988, 1997b). At pres- ent, our field data indicate that E. janetscheki does not meet two of these premises: the spe- cies was not collected on or above the soil surface and it had no distinct reproductive pe- riod. A predominance of females over males was reported for the European palpigrad species Eukoenenia mirabilis (Gruner 1993; Conde 1996). In E. janetscheki almost twice as many females as males were captured in the primary upland forest at Reserva Ducke. The number of females in three species of Symphyla from the same forest and from the secondary up- land forest at Rio Taruma Mirfm was even 2- 4 times higher than of males (Adis et al. 1997b). Predominance of females assures the continuation of a species, and it probably in- creases the chance to locate spermatophores deposited by males in the soil. ACKNOWLEDGEMENTS This study was supported by a grant from the Max-Planck-Society for the second author. We wish to acknowledge the valuable support received by PD Dr. W.J. Junk, Head of the Tropical Ecology Working Group at the Max- Planck-Institute (MPI) for Limnology in Plon, Germany. Dr. James Carico (Lynchburg/ USA), Dr. Norman Platnick (New York/USA), Dr. Sergei I. Golovatch (Moscow/Russia) and Prof. Dr. Volker Mahnert (Geneve/Switzer- land) are thanked for valuable comments on the manuscript. Berit Hansen (MPI Plon) is thanked for making the drawings. Dr. Johann Bauer, MPI for Biochemistry (Martinsried/ Germany) kindly provided the backsearch of literature on Palpigradi, covering the last 30 years. LITERATURE CITED Adis, J. 1987. Extraction of arthropods from neo- tropical soils with a modified Kempson appara- tus. J. Trop. Ecol. 3(2):131-138. Adis, J. 1992. Uberlebensstrategien terrestrischer Invertebraten in Uberschwemmungswaldern Zentralamazoniens. Verb, naturwiss. Ver. Ham- burg (NF), 33:21-114. Adis, J. 1997. Terrestrial invertebrates: Survival strategies, group spectrum, dominance and activ- ity patterns. Pp. 299-317. In The Central Ama- zon floodplain. Ecology of a pulsing system. (W.J. Junk, ed.). Ecological Studies 126. Spring- er, Heidelberg. Adis, J. & M. Latif. 1996. Amazonian arthropods react to El Nino. Biotropica, 28(3):403-408. Adis, J. & M.O. de A. Ribeiro. 1989. Impact of deforestation on soil invertebrates from Central Amazonian inundation forests and their survival strategies to long-term flooding. Water Quality Bull., 14(2):88-98, 104. Adis, J. & H.O.R. Schubart. 1984. Ecological re- search on arthropods in Central Amazonian for- est ecosystems with recommendations for study procedures. Pp. 111-144. In Trends in ecolog- ical research for the 1980s. NATO Conference Series, Series I. (J.H. Cooley & EB. Golley, eds.). Ecology, Vol. 7. Plenum Press, New York, London. 344 pp. Adis, J., Y.D. Lubin & G.G. Montgomery. 1984. Arthropods from the canopy of inundated and terra firme forests near Manaus, Brazil, with critical considerations on the pyrethrum-fog- ging technique. Stud. Neotrop. Fauna Environ., 19:223-236. Adis, J., J.W de Morais & H.G. de Mesquita. 1987a. Vertical distribution and abundance of ar- thropods in the soil of a Neotropical secondary forest during the rainy season. Stud. Neotrop. Fauna Environ., 22(4): 189-197. Adis, J., J.W. de Morais & E.E de Ribeiro 1987b. Vertical distribution and abundance of arthropods in the soil of a Neotropical secondary forest dur- ing the dry season. Trop. Ecol. 28(1): 174-181. Adis, J., J.W. de Morais, E.E Ribeiro & J.C. Ri- beiro. 1989a. Vertical distribution and abun- dance of arthropods from white sand soil of a Neotropical campinarana forest during the rainy season. Stud. Neotrop. Fauna Environ., 24(4): 193-200. Adis, J., V Mahnert, J.W de Morais & J.M.G. Ro- drigues. 1988. Adaptation of an Amazonian 332 THE JOURNAL OF ARACHNOLOGY pseudoscorpion (Arachnida) from dryland forests to inundation forests. Ecology, 69(1):287-291. Adis, J., W. Paarmann, C.R. Fonseca & J.A. Rafael. 1997a. Knock-down efficiency of natural pyre- thrum and survival rate of arthropods obtained by canopy fogging in Central Amazonia. Pp.67- 81. /« Canopy Arthropods. (N.E. Stork, J. Adis & R.K. Didham, eds.). Chapman & Hall, Lon- don. 576 pp. Adis, J., E.F. Ribeiro, J.W. de Morais & E.T.S. Cav- alcante. 1989b. Vertical distribution and abun- dance of arthropods from white sand soil of a Neotropical campinarana forest during the dry season. Stud. Neotrop. Fauna Environ., 24(4): 201-211. Adis, J., U. Scheller, J.W. de Morais, C. Rochus & J.M.G. Rodrigues. In press. Amazonian Sym- phyla (Myriapoda) from non-flooded upland for- ests and their adaptations to inundation forests. Entmol. Scandinavica Suppl. Adis, J., A. Minelli, J.W. de Morais, L.A. Pereira, F. Barbieri & J.M.G. Rodrigues. 1997c. On abundance and phenology of Geophilomorpha (Chilopoda) from Central Amazonian upland for- ests. Ecotropica, 2(2): 165-175. Broza, M., D. Poliakov & B. Conde. 1993. The first record of the order Palpigradida (Arachnida) in Israel and the occurrence of arachnids in soils of Mediterranean pine forests. Israeli J. ZooL, 39:147-151. Cavalli-Sforza, L. 1972. Grundziige biologisch- medizinischer Statistik. G. Fischer, Stuttgart. 212 pp. Conde, B. 1984a. Les Palpigrades: quelques as- pects morphobiologiques. Rev. ArachnoL, 5(4): 133-143. Conde, B. 1984b. Palpigrades (Arachnida) d’ Europe, des Antilles, du Paraguay et de Thai- lande. Rev. Suisse ZooL, 91(2):369-391. Conde, B. 1986. Les palpigrades du nouveau mon- de: etat des connaissances. Mem. Soc. R. Beige Entomol., 33:67-73. Conde, B. 1993. Description du male de deux es- peces de Palpigrades. Rev. Suisse ZooL, 100(2): 279-287. Conde, B. 1996. Les palpigrades, 1885-1995: Aquisition et lacunes. Rev. Suisse ZooL, voL hors sme 1:87-106. Conde, B. 1997. Description complementaire du Palpigrade Bresilien Eukoenenia janetscheki Conde. Amazoniana, 14(3/4): In press. Gruner, H.-E. (Ed.). 1993. Lehrbuch der Speziellen Zoologie. Bd. I: Wirbellose Tiere. 4. Teil: Ar- thropoda (ohne Insekten). G. Fischer, Stuttgart. 1279 pp. Harada, A.Y. & A.G. Bandeira. 1994a. Estratifi- cagao e densidade de invertebrados em solo ar- enoso sob floresta e plantios arboreos na Ama- zonia central durante a estagao seca. Acta Amazonica, 24(1/2): 103-1 18. Harada, A.Y. & A.G. Bandeira. 1994b. Estratifi- cagao e densidade de invertebrados em solo ar- giloso sob floresta e plantios arboreos na Ama- zonia central durante a estagao seca. Bol. Mus. Par. Emflio Goeldi, ser. ZooL, 10(2):235-25L Janetschek, H. 1957. Das seltsamste Tier Tirols. Palpenlaufer (Arachn., Palpigradida): Stellung, Verbreitung, Arten, Bibliographie. Pp. 192-214, In Kufsteiner Buch Bd. Ill (Schlem-Schr. Bd. 158). Univ. Verlag Wagner, Innsbruck, 223 pp. Mahnert, V. & H. Janetschek. 1970. Bodenlebende Palpenlafer in der Alpen (Arachnida, Palpigrad- ida). Oecologia, 4(1): 106-1 10. Morais, J.W. de. 1985. Abundancia e distribuigao vertical de Arthropoda do solo numa floresta pri- maria nao inundada. M.Sc. thesis, CNPq/INPA/ FUA. Manaus, Brazil. 92 pp. Penny, N.D. & J, Arias. 1982, Insects of an Am- azon forest. Columbia Univ. Press, New York. 269 pp. Price, D.W, 1975. Vertical disribution of small ar- thropods in a California pine forest soil. Ann. Entomol, Soc, America, 68(1): 174-1 80. Ribeiro, M.O. de A. 1994. Abundancia, distribui- gao vertical e biomassa de artropodos do solo em uma capoeira na Amazonia Central. M.Sc. thesis, INPA/UFAM. Manaus, Brazil. 106 pp. Ribeiro, M. de N.G. & J. Adis. 1984. Local rainfall variability — a potential bias for bioecological studies in the Central Amazon. Acta Amazonica, 14(1/2):159-174. Rodrigues, J.M.G. 1986. Abundancia e distribui- gao vertical de Arthropoda do solo, em capoeira de terra firme. M.Sc. thesis, CNPq/INPA/FUA. Manaus, Brazil. 80 pp. Serafino, A. & J.F. Merino. 1978. Poblaciones de microartropodos en diferentes suelos de Costa Rica. Rev. Trop. Biol., 26(1):139-15L Manuscript received 12 November 1996, accepted 25 February 1997. 1997. The Journal of Arachnology 25:333-351 GROUND-LAYER SPIDERS (ARANEAE) OF A GEORGIA PIEDMONT FLOODPLAIN AGROECOSYSTEM: SPECIES LIST, PHENOLOGY AND HABITAT SELECTION Michael L. Draney^: Department of Entomology, University of Georgia, Athens, Georgia 30602 USA ABSTRACT. Monthly pitfall trapping in 1990 and 1991 at Horseshoe Bend Experimental Area, Clarke County, Georgia, yielded 112 species of spiders belonging to 25 families. Examination of additional collections brings the site total to 145 species in 26 families, including southern or southeastern range extensions for Agelenopsis kastoni, Sphodros atlanticus, Bathyphantes pallidus, Eridantes erigonoides, Floricomus tallulae, Grammonota inomata, and Walckenaeria Carolina, and a northeastern range exten- sion for Paratheridula pemiciosa. Ceraticelus emertoni and Neriene redacta are also reported from Geor- gia for the first time. The proportional distribution of pitfall-trapped species within families does not differ significantly from that reported for Berry’s (1966) pitfall trapping in the North Carolina Piedmont (about 450 km away), suggesting regional similarity of the Piedmont ground-layer spider fauna. Data on phe- nology and relative catch of species among the four habitats sampled (conventional and no-tillage agri- cultural fields, grassy field borders, and the surrounding deciduous riparian forest) are given for the most abundant species. Habitat selection of 15 abundant species was statistically analyzed; most of the species’ populations displayed strong preferences for particular habitats. It is clear that species “spillover” from adjacent habitats contributes to the faunal richness of each habitat, and that maintenance of a mosaic of habitats within an agroecosystem landscape maximizes spider biodiversity. Since Chamberlin & Ivie’s (1944) seminal effort, little work has been conducted on the ground-layer spider fauna of the southeastern Piedmont Plateau region, the mid-elevation area located between the Appalachian Moun- tains and the Atlantic Coastal Plain. One no- table exception is Berry’s (1966, 1970) study of the old-field succession of the North Car- olina Piedmont, which lists 331 species from the region, including 217 (66%) from over 10,000 pitfall trap/days. The present work re- ports on the spiders collected during ecologi- cal research conducted at Horseshoe Bend Ex- perimental Area, a mosaic of agricultural plots and forest on the floodplain of the Oconee River on the Georgia Piedmont. In order to better understand the distribu- tion of spiders within the various habitats of this agroecosystem, systematic pitfall trapping was conducted in four distinct (but adjacent) habitats: (1) the natural floodplain forest, un- disturbed by management practices, (2) con- * Current address: Savannah River Ecology Labo- ratory, Drawer E, Aiken, South Carolina 29802 USA ventional tillage agricultural fields, (3) no-till- age agricultural fields, and (4) the grassy field borders that surround these habitats. Although the four habitat types at Horseshoe Bend are rather small and in close proximity (all within 10 m of each other), they are typical of the modem fragmented landscape of Georgia. Much of the cultivated land in the Georgia Piedmont consists of small plots with a high proportion of “edge” (Turner & Ruscher 1988). The Horseshoe Bend agricultural fields are planted in grain sorghum in the summer and in winter-rye and crimson clover in the winter. Blumberg (1979) examined ground-layer spi- ders in these systems at Horseshoe Bend as part of an overall arthropod community char- acterization, but the low sampling intensity and broad scope of the study meant that the spider assemblages were not extensively char- acterized and analyzed. The only other study of spiders in grain sorghum is Bailey & Cha- da’s (1968) work describing assemblages in Oklahoma. Blumberg (1979) and this study remain the only examinations of grain sor- 333 334 THE JOURNAL OF ARACHNOLOGY ghum spider assemblages in the southeastern United States. The Horseshoe Bend site might be expected to harbor a fairly high diversity of species. The floodplain on which the site is located is a mesic, highly productive site, and the fairly undisturbed floodplain forest is now a rather uncommon habitat in the Georgia Piedmont. The north-facing slopes of the Oconee River harbor a flora (e.g., beech, Fagus grandifold) more characteristic of forests farther north (F.B. Golley pers. comm.). The various open field habitats increase the site’s spatial hetero- geneity and plant species diversity. Finally, ri- parian zones such as Horseshoe Bend’s flood- plain forest may serve as corridors, allowing organisms that are more eurytopic in the mountains or the coastal plain to extend their ranges into the Piedmont. The objectives of this work are to list the spiders that occur in various habitats on a Georgia Piedmont floodplain; to compare the resultant data with Berry’s (1966) list of North Carolina Piedmont fauna, and with other fau- nal lists compiled using similar collecting methods; and to present phenology and habitat selection data for some of the most abundantly trapped species at the site. METHODS Study site and habitats. — This study was conducted at the University of Georgia’s Horseshoe Bend Experimental Area near Ath- ens in Clarke County, Georgia (33°55'52"N, 83°21'04'W). It is located on the floodplain of the Oconee River (elevation 244 m) and much of the 14 ha site is occupied by a deciduous riparian forest. The 2 ha currently occupied by the agricultural plots were used as pasture pri- or to 1964. Between 1964-1978, studies of old-field succession occupied the site (Blum- berg & Crossley 1983). The area is relatively flat (slopes < 3%) and the soil is a well- drained, moderately acidic sandy-clay loam (House & Parmelee 1985). The area is flooded in certain years; one comer of the study area was flooded during a week in the winter of 1990, several months before this study was conducted. The four habitats sampled are as follows: Floodplain forest: This habitat is a decid- uous forest typical of southeastern Piedmont riparian zones. Some areas have not been logged for a considerable period, probably in excess of 100 years, judging from the diam- eter of some of the trees (F. Golley pers. comm.), though some of the forest was logged as recently as 50 years ago (P. Hendrix pers. comm.). Dominant trees include sweetgum (Liquidambar styraciflua), tulip poplar (Liri- odendron tulipifera), white oak {Quercus alba), water oak (Quercus nigra), chestnut oak {Quercus prinus) and beech {Fagus gran- difola). The dominant understory tree is flow- ering dogwood {Comus florida), and poison ivy {Toxicodendron radicans) is abundant in the herbaceous layer. Grassy field border: The agricultural fields are separated from the forest by grassy field borders approximately 5 m wide. These con- sist mainly of fescue grass {Festuca sp.) in- terspersed with annuals. The field borders are contiguous with larger areas of meadow of up to 15 m width in other areas of the Horseshoe Bend clearing. All meadow areas are main- tained by periodic mowing, usually four times during the growing season. No-tillage agroecosystem: Four of the eight 32 X 32 m experimental sub-plots on the site have been maintained as no-tillage agricultur- al plots since 1978. Sorghum {Sorghum bi- color) is grown as a summer crop (approx. June-October), and winter rye {Secale cereale) and crimson clover {Trifolium incar- natum) are grown as cover crops in the winter (approx. November-May). Major weeds in- clude pigweed {Amaranthus retroflexus), sick- lepod {Cassia obtusifolia), and Johnson grass {Sorghum halepense) (Parmelee et al. 1989). At the end of both growing seasons the crop is harvested and the next crop is planted by drilling (summer crop) or by surface broad- casting (winter crop). Lack of disturbance to the soil allows a thick litter layer to build up, creating a very different ground-layer micro- cfimate than in the conventional tillage plots (Hendrix et al. 1986). Conventional tillage agroecosystem: The four conventional tillage plots are maintained under the same crop rotation as are the no- tillage plots. However, after the crops are har- vested, the conventional tillage plots are moldboard plowed, disked, rotary tilled, and seeded. At the beginning of each growing sea- son, the conventional tillage plots are essen- tially bare, exposed soil. Conventional tillage plots are thus the most highly disturbed of the four floodplain habitats, with the forest being DRANEY—PffiDMONT FLOODPLAIN SPIDERS the least disturbed by management practices. No pesticides or irrigation are used on any habitat. For specific dates of plowing, plant- ing, and mowing, see Draney (1992). Sampling. — ^The ground-layer spider as- semblages of the four habitats were sampled with 9.5 cm diameter plastic pitfall traps (Morrill 1975) containing 70% ethanol. Traps were run only during days without significant precipitation. Pitfall traps were run for one 24- hour period approximately once a month from August 1990 to August 1991. Twenty-four hour trapping periods sample fauna equally during all times of the day, avoiding the diel bias that can result in a distorted view of com- munity structure (Costa & Crossley 1991). During the first five trapping periods (August 1990-January 1991), ten pitfall traps were placed in each habitat. To obtain a large enough sample to examine the patterns of spi- der diversity in the four habitats, the number of pitfall traps in each habitat was increased to 20, starting in February 1991 and continu- ing for the duration of sampling. Cumulative sampling effort was 760 trap-days, 190 days per habitat. For specific trapping dates, see Draney (1992). Traps were placed in lines of five traps each, resulting in a stratified-random design. Traps within a line were separated by approx- imately 5 m. Lines were separated from each other by a randomly-selected distance be- tween 5-15 m. In the agroecosystems and the forest, the first trap of each line was placed 5 m from the habitat boundary and lines contin- ued perpendicular to the habitat boundary. In the grassy field margins, traps were placed ap- proximately in the center of the field margin strips, which were 4-5 m wide. For purposes of data analyses, each five trap pitfall line was pooled as a sample unit. Processing of samples.— All Arachnida (other than Acari) were removed from the samples by visual inspection under a dissect- ing microscope, and stored in clean 70% eth- anol for subsequent identification. Initially, spiders in each sample were sexed (male, fe- male, or immature) and identified to morpho- species. Subsequently, animals were identified to species. Errors in initial morphospecies as- signment precluded analysis of phenology and habitat selection for species in certain genera, including Drassyllus, Gnaphosa, Scotinella, Theridion, Meioneta, and two species of 335 Phrurotimpus {P. borealis and P. emertoni). Since specimens in many of the original sam- ples were removed for use as voucher mate- rial, accurate re-examination of the original samples was not possible. Other sources of material. — In order to include as much of the site’s spider fauna as possible, all available material collected from Horseshoe Bend was examined in compiling the species list, although only the above-men- tioned pitfall data were used in analyses of phenology and habitat selection. The sources are listed in Table 1. All material was determined by the author, 1992-1995, except as noted in the acknowl- edgments. Identifications were confirmed dur- ing visits to the National Museum of Natural History and the American Museum of Natural History, 1995-1996. Voucher specimens of all taxa have been deposited at the University of Georgia’s Natural History Museum. Comparison of pitfall faunas. — Barnes & Barnes (1955) remains the most comprehen- sive comparison of southeastern spider assem- blages. Their work described the “abstract” spider community which occurs fairly con- stantly in the broomsedge successional habi- tats occurring on the southeastern Piedmont, and was the first paper to identify such a pre- dictable spider assemblage (Turnbull 1973; Foelix 1982). Comparison of the pitfall fauna of the present study with that of Berry (1966) could indicate the degree of constancy in the ground-layer Piedmont fauna, at least between two widely separated sites (about 450 km apart) in the region. Differences in samphng effort and nomen- clatural changes in the years since Berry’s (1966) work make a direct species-level com- parison impossible. However, the faunas can be compared at the level of family by exam- ining the proportion of total species found in each family (Table 2). Family richness (28 vs. 24) varies little between the lists. If forest and field assemblages are similar in structure, function, and biogeographic history through- out the Piedmont region (the abstract com- munity sensu Barnes & Barnes 1955), then given families should likewise either be di- verse and dominate assemblages in terms of species, or remain species-poor throughout the region. To test this, I compared the proportion of total species found in each family at Horse- shoe Bend with the number of species ex- 336 THE JOURNAL OF ARACHNOLOGY Table 1. — Spiders of Horseshoe Bend Experimental Area, Clarke County, Georgia, All species were trapped in the trough-type or monthly cup-type pitfall traps run in 1990 and 1991 except those species marked with an * *, which are not considered “pitfall species” and not included in comparisons with other pitfall trap faunas. Sources of specimens are: 1, Collected by J. L Richardson in 1967. Collection methods not known; 2, Collected by G. Bakelaar, 1975, via vacuum sampling and/or sweepnetting of herbaceous vegetation; 3, Hand collected or observed by M. Draney at various times in 1990 and 1991; 4, Large formalin-filled, trough-like directional pitfall traps (140 X 40 cm) placed at habitat boundaries and operated 26 May-8 July, 1991; 5, Monthly 24-hour cup-type pitfall traps in 1990 and 1991 (total effort = 918 trap/days). prob. = “probably;” adult specimens/male specimens needed to confirm identification, cf. = “near;” specimen may be an individual of that species, but differences from the descriptions indicate that it may belong to a closely related species. Genera within families and families within suborders are listed alphabetically. Nomenclature follows Platnick (1993). Source of specimens Mygalomorphae Atypidae Sphodros atlanticus Gertsch & Platnick 1980 4 Ctenizidae Ummidia audouini (Lucas 1835) 4 Cyrtaucheniidae Myrmekiaphila fluviatilis (Hentz 1850) 3, 5 Araneomorphae Agelenidae Agelenopsis kastoni Chamberlin & Ivie 1941 2, 3, 5 Amaurobiidae Coras medicinalis (Hentz 1821) 4 Coras sp. 4 Wadotes bimucronatus (Simon 1898) 4, 5 Anyphaenidae *Teudis mordax (O. P.-Cambridge 1896) 2 *Wulfila saltabunda (Hentz 1847) 2 Araneidae Acacesia hamata (Hentz 1847) 2, 4, 5 *Acanthepeira stellata (Walckenaer 1805) 2, 3 "^Araneus bicentenarius (McCook 1888) 3 *Araneus sp. 2 ^Araniella displicata (Hentz 1847) 2 *Argiope aurantia Lucas 1833 2, 3 *Cyclosa turbinata (Walckenaer 1841) 1, 2, 3 Gea heptagon (Hentz 1850) 1, 4, 5 *Larinia directa (Hentz 1847) 2 *Micrathena gracilis (Walckenaer 1805) 3 Micrathena mitrata (Hentz 1850) 5 *Micrathena sagittata (Walckenaer 1841) 2 ^Neoscona arabesca (Walckenaer 1841) 2, 3 Clubionidae Cheiracanthium inclusum (Hentz 1847) 2, 5 Clubiona sp. A 1, 2, 5 *Clubiona sp. B 2 Elaver prob. exceptus (L. Koch 1866) 4 Corinnidae Castianeira cingulata (C. L. Koch 1841) 3, 5 Castianeira gertschi Kaston 1945 4, 5 Castianeira longipalpa (Hentz 1847) 4, 5 Castianeira trilineata (Hentz 1847) 4 Trachelas deceptus (Banks 1895) 2, 4, 5 DRANEY— PffiDMONT FLOODPLAIN SPIDERS 337 Table 1. — Continued. Source of specimens Ctenidae Anahita punctulata (Hentz 1844) 5 Dictynidae Dictyna volucripes Keyserling 1881 2, 5 Gnaphosidae *Callilepis sp. 2 Cesonia bilineata (Hentz 1847) 2, 4 Drassyllus covensis Exline 1962 5 Brassy llus dixinus Chamberlin 1922 4, 5 Drassyllus eremitus Chamberlin 1922 4 Drassyllus ellipes Chamberlin & Gertsch 1940 4 Drassyllus novus (Banks 1895) 5 Gnaphosa fontinalis Keyserling 1887 4, 5 Gnaphosa sericata (L. Koch 1866) 3, 4, 5 Sergiolus ocellatus (Walckenaer 1837) 1, 4 Zelotes aiken Platnick & Shadab 1983 3, 4, 5 Zelotes duplex Chamberlin 1922 4 Hahniidae Neoantistea agilis (Keyserling 1887) 5 Neoantistea riparia (Keyserling 1887) 5 Linyphiidae Erigoninae Ceraticelus emertoni (O. P.-Cambridge 1874) 1, 5 Ceratinella brunnea Emerton 1882 5 Ceratinops crenatus Emerton 1882 5 Eperigone fradeorum (Berland 1932) 4, 5 Eperigone inomata Ivie & Barrows 1935 5 Eridantes erigonoides (Emerton 1882) 4, 5 Erigone autumnalis Emerton 1882 1, 4, 5 Floricomus tallulae Chamberlin & Ivie 1944 5 Floricomus sp.? 5 Goneatara platyrhinus (Crosby & Bishop 1927) 5 Grammonota inornata Emerton 1882 5 Idionella sclerata (Ivie & Barrows 1935) 5 Walckenaeria Carolina Millidge 1983 5 Walckenaeria spiralis (Emerton 1882) 4, 5 Linyphiinae Bathyphantes pallidus (Banks 1892) 4, 5 Centromerus latidens (Emerton 1882) 5 Florinda coccinea (Hentz 1850) 1, 3, 4, 5 Frontinella pyramitela (Walckenaer 1841) 1, 3, 4, 5 Lepthyphantes sabulosus (Keyserling 1886) 5 Meioneta angulata (Emerton 1882) 5 Meioneta barrow si Chamberlin & Ivie 1944 5 Meioneta cf. leucophora Chamberlin & Ivie 1944 5 Meioneta cf. longipes Chamberlin & Ivie 1944 5 Meioneta micaria (Emerton 1882) 5 Meioneta cf. meridionalis (Crosby & Bishop 1936) 5 Meioneta serrata (Emerton 1909) 5 Neriene radiata (Walckenaer 1841) 5 Neriene redacta Chamberlin 1925 1, 5 Neriene variabilis (Banks 1892) 5 Tennesseelum formicum (Emerton 1882) 4, 5 338 THE JOURNAL OF ARACHNOLOGY Table 1. — Continued. Source of specimens Liocranidae Agroeca prob. pratensis Emerton 1890 5 Phrurotimpus alarms (Hentz 1847) 4, 5 Phrurotimpus borealis (Emerton 1911) 3, 4, 5 Phrurotimpus emertoni (Gertsch 1935) 4, 5 Scotinella fratrella (Gertsch 1935) 5 Scotinella redempta (Gertsch 1941) 5 Lycosidae Allocosa funerea (Hentz 1844) 3, 4, 5 Gladicosa gulosa (Walckenaer 1837) 1, 5 Hogna lenta (Hentz 1844) 2, 3, 5 Hogna timuqua (Wallace 1942) 3, 4, 5 Pardosa atlantica Emerton 1913 2, 3, 4, 5 Pardosa milvina (Hentz 1844) 4, 5 Pardosa pauxilla Montgomery 1904 4, 5 Pirata iviei Wallace & Exline 1978 1, 4, 5 Rabidosa rabida (Walckenaer 1837) 3, 4, 5 Schizocosa ocreata (Hentz 1844) 4, 5 Schizocosa prob. bilineata (Emerton 1885) 5 Oxyopidae *Oxyopes aglossus Chamberlin 1929 2 Oxyopes salticus Hentz 1845 1, 2, 3, 4, 5 Peucetia viridans (Hentz 1832) 2, 3, 4 Philodromidae *Philodromus imbecillus Keyserling 1880 2 Philodromus sp. A 2 *Thanatus formicinus (Clerck 1757) 1 Tibellus duttoni (Hentz 1847) 1, 4 Pisauridae Pisaurina mira (Walckenaer 1837) 3, 5 Salticidae Corythalia canosa (Walckenaer 1837) 4, 5 Habrocestum parvulum (Banks 1895) 4, 5 Habronattus coecadus (Hentz 1846) 2, 5 ^Maevia inclemens (Walckenaer 1837) 2 Marpissa lineata (C. L. Koch 1848) 4 ^Marpissa pikei (Peckham & Peckham 1888) 2 *Metaphidippus galathea (Walckenaer 1837) 2 Metaphidippus sexmaculatus (Banks 1895) 2, 4, 5 Phidippus audax (Hentz 1845) 2, 3, 4 *Phidippus princeps (Peckham & Peckham 1883) 2 * Phidippus rimator (Walckenaer 1837) 1, 2 *Sarinda hentzi (Banks 1913) 1 Sitticus cursor Barrows 1919 4, 5 Sitticus prob. magnus Chamberlin & Ivie 1944 5 Thiodina puerpura (Hentz 1846) 2, 4 Zygoballus sexpunctatus (Hentz 1845) 2, 4 Segestriidae Ariadna bicolor (Hentz 1842) 5 T etragnathidae Glenognatha foxi (McCook 1893) 3, 4, 5 Pachygnatha tristriata C. L. Koch 1845 4 Tetragnatha laboriosa Hentz 1850 2, 3, 4 Tetragnatha straminea Emerton 1884 2, 4 DRANEY— PffiDMONT FLOODPLAIN SPIDERS 339 Table 1. — Continued. Source of specimens Theridiidae Argyrodes lacerta (Walckenaer 1841) 5 Dipoena nigra (Emerton 1882) 5 Latrodectus mactans (Fabricius 1775) 3, 5 Paratheridula perniciosa (Keyserling 1886) 5 Pholcomma hirsutum Emerton 1882 5 Phoroncidia americana (Emerton 1882) 5 Steatoda americana (Emerton 1882) 5 Stemmops omatus (Bryant 1933) 5 Theridion (2-3 spp.) 5 Theridula opulenta (Walckenaer 1841) 1, 2, 4 Thomisidae *Misumena vatia (Clerck 1757) 1, 2 *Misumenoides formosipes (Walckenaer 1837) 1, 2 *Misumenops (2 spp.) 2 *Synema parvulum (Hentz 1847) 1 *Tmarus angulatus (Walckenaer 1837) 2 Xysticus ferox (Hentz 1847) 1, 2, 3, 4, 5 Xysticus triguttatus Keyserling 1880 2, 4, 5 Xysticus sp. 4, 5 Uloboridae Uloborus glomosus (Walckenaer 1841) 3, 4 Zoridae Zora pumila (Hentz 1850) 4 pected in each family based on the propor- tional representation of the Berry (1966) data via a Chi-square test (alpha 0.05; Table 3). For both lists, species identified only to genus were included only if no congener exists in the same list. Placement of species within families follows Platnick (1993) rather than Berry’s (1966) original placement. In order to account for rare families that were not present in both lists, I lumped species from all fami- lies representing < 5% of species richness of the Berry (1966) data into a single “other families” category. If there is a similarity of ground-layer fau- nas throughout the Piedmont region, it is ex- pected that the structure of the Horseshoe Bend fauna would be more similar to the Piedmont fauna of Berry (1966) than to pitfall fauna of other regions. I examined this by comparing the fauna of the present study to six other complete lists of pitfall spider spe- cies from outside the Piedmont Plateau region (Table 3) using the same chi-square test pro- cedure. Data analysis. “=-In comparing the Horse- shoe Bend fauna with other faunas, only the pitfall samples are included, due to the un- quantifiable and uneven collecting of vegeta- tion-layer spiders. The 1990-91 pitfall collec- tions (cup and trough traps) together represent a significant portion of the total spider diver- sity sampled at the site, including 112 species (77% of Horseshoe Bend total) belonging to 25 families, of which 71 species (63% of pit- fall fauna) were sampled only with these methods. For examining phenological patterns, data representative of the entire year without tem- poral bias are preferable. The phenology data set consists of 10 traps per habitat-date for all dates, August 1990-August 1991, and in- cludes 960 adult spiders trapped over 480 trap/days. To ensure taxonomic accuracy, only adult spiders were included in examining species’ habitat preferences. In order to maximize sam- ple size while avoiding temporal bias in sam- pling effort, only months with 20 pitfall traps (February- August 1991) were included in the data set from which habitat selection infor- mation was extracted. Each habitat was sam- pled with four 5-pitfall sample units on each of seven dates, giving 28 samples at each hab- 340 THE JOURNAL OF ARACHNOLOGY Table 2. — Comparison of Horseshoe Bend pitfall fauna with North Carolina Piedmont pitfall fauna listed in Berry (1966). Only species captured in pit- falls in the piedmont are recorded for Berry (1966). Taxa identified only to “sp.” were included only if no congener was listed. Families are listed in de- scending order of species richness of the Horseshoe Bend fauna, with ties listed alphabetically. Family Number of species (Ber- (this ry study) 1966) % in this study % in Berry 1966 Linyphiidae 29 47 25.89 21.66 Gnaphosidae 11 17 9.82 7.83 Lycosidae 11 34 9.82 15.67 Salticidae 10 23 8.93 10.60 Theridiidae 10 12 8.93 5.53 Liocranidae 6 6 5.36 2.76 Corinnidae 5 3 4.46 1.38 Tetragnathidae 4 4 3.57 1.84 Araneidae 3 16 2.68 7.37 Clubionidae 3 2 2.68 0.92 Thomisidae 3 11 2.68 5.07 Amaurobiidae 2 2 1.79 0.92 Hahniidae 2 4 1.79 1.84 Oxyopidae 2 4 1.79 1.84 Agelenidae 1 4 0.89 1.84 Atypidae 1 1 0.89 0.46 Ctenidae 1 0 0.89 0.00 Ctenizidae 1 1 0.89 0.46 Cyrtaucheniidae 1 1 0.89 0.46 Dictynidae 1 5 0.89 2.30 Philodromidae 1 5 0.89 2.30 Pisauridae 1 3 0.89 1.38 Segestriidae 1 1 0.89 0.46 Uloboridae 1 0 0.89 0.00 Zoridae 1 1 0.89 0.46 Anyphaenidae 0 5 0.00 2.30 Mimetidae 0 1 0.00 0.46 Mysmenidae 0 1 0.00 0.46 Nesticidae 0 1 0.00 0.46 Oonopidae 0 1 0.00 0.46 Titanoecidae 0 1 0.00 0.46 Total species 112 217 100.00 100.00 Total families 25 29 itat. This data set consists of 1436 adult spi- ders trapped over 560 trap/days. Data from each of 15 species in which at least 20 adults were trapped were analyzed separately by 2-Way ANOVA, with habitat as the major predictive variable and blocked by sample date. Data showing a significant among-habitat effect were subjected to a uni- variate ANOVA and habitat means separation via Fisher’s LSD. RESULTS AND DISCUSSION The Horseshoe Bend spider fauna.^ — In all, 145 spider species belonging to 26 fami- lies have been collected at Horseshoe Bend (Table 1). This list represents the most exten- sive pitfall trapping survey yet conducted on the Georgia Piedmont. Note, however, that species collected only in 1967 and/or 1975 should be viewed with caution, as the collec- tions were made in old field successional hab- itats that are largely absent from the site today. A Chi-square test showed that the observed proportional distribution of species within families was not significantly different from the distribution predicted based on the Berry (1966) list (Table 3). Thus, the two faunas have similar family-level structure, which is consistent with the concept of an abstract Piedmont ground-layer assemblage. In contrast to the Piedmont fauna compar- ison, the species-within-families distribution of the Horseshoe Bend fauna was significantly different (alpha = 0.05) from each of the six non-Piedmont faunas (Table 3). While the above does not constitute a rigorous test of the hypothesis that there exists an “abstract Pied- mont ground-layer spider assemblage”, it is at least consistent with such a hypothesis, and suggests some broad regional similarity of ground-layer spider faunas at the level of fam- ily. Range extensions.— The pitfall data in- clude records of new range extensions for eight species. Seven of these are southern or southeastern and one is a northeastern range extension. The predominance of southern over northern range extensions at this site is not surprising considering: 1) the site is located near the southeastern comer of the continent, so much more land occurs to the north and west of this site than to the south and east, and 2) much more spider collecting has been conducted to the north of this area, due to the historical distribution of arachnologists in the midwest and middle and northern Atlantic states. Agelenidae: Agelenopsis kastoni Chamber- lin & Ivie 1941: Two males were captured in the forest on 26-27 March and another male on 23-24 April 1991. Few collection localities of this spider have been published since DRANEY— PffiDMONT FLOODPLAIN SPIDERS 341 Table 3. — Results of Chi-square test (alpha = 0.05) of hypothesis that distribution of pitfall spider species richness between families is similar between present study and other faunas. Comparison studies are listed in descending order by number of families. Study Location Habitats Present study Georgia Piedmont Riparian fields and forest Berry 1966 North Carolina Piedmont Forests and old fields Muma 1973 Central Florida Pine, citrus, residential Bailey & Chada 1968 Oklahoma Grain sorghum Muma 1975 New Mexico Desert grassland, sand dunes Maelfait & DeKeer 1990 Belgium Grazed pasture, grassy edge Muma & Muma 1949 Nebraska Prairie, wooded ravine Koponen 1992 Northwest Territories, Canada Various low arctic habitats Chamberlin & Ivie (1941) described the spe- cies from single male and females types from Haddam, Connecticut. It is known from Oco- nee and Pickens Counties, South Carolina (Gaddy & Morse 1985), and was listed in Ber- ry (1966) as a North Carolina Piedmont pitfall spider, also collected in forest. The Horseshoe Bend records extend the known range of the species at least 50 km south. Recently, four males were trapped on the inner coastal plain as well, extending the known range even far- ther south (South Carolina: Barnwell County, Savannah River Site; Set-Aside #29, Scrub Oak Natural Area, 17 April-3 May 1996. ColL/Det. M. Draney). Atypidae: Sphodros atlanticus Gertsch & Platnick 1980: One male was captured in a trough trap between the forest and the grassy field border during the last week of June 1991. Another male was trapped one week later at the edge of the sorghum field, about 75 m from the forest edge, where the spider pre- sumably originated. Hall County, Georgia is the previous southernmost collection record; these specimens extend the known range of the species southward by about 40 km. Other localities reported for S. atlanticus are Jackson County, Illinois; Carteret and Jackson Coun- ties, North Carolina; and Spotsylvania Coun- ty, Virginia (Gertsch & Platnick 1980; Coyle et al. 1985). Berry’s (1966, 1970) list does not include S. atlanticus but lists Sphodros niger (Emerton) (listed as Atypus)\ however, like many of the taxa on the present list, Atypidae was revised and S. atlanticus described since the publication of Berry’s (1966, 1970) list (Gertsch & Platnick 1980). Linyphiidae: Bathyphantes pallidus (Banks 1892): Seven adult individuals of this species were captured in the no-tillage and grassy field border habitats in March, May, June, July and September 1991. The species is widely distributed across the United States to about 34° N, with the southernmost localities at Highlands and Clingman’s Dome, North Car- olina (Ivie 1969). The Horseshoe Bend re- cords extend the known range of the species southward by about 120 km. However, a sin- gle female was recently trapped even further south on the inner coastal plain (South Caro- lina: Aiken County; Jackson. Deciduous woods behind 110 Cowden St.; Pitfall, 12-16 March 1995. Coll./Det. M. Draney). These are the southernmost records for the genus, except that an undetermined species of Bathyphantes was reported from Florida (Anonymous 1986). Linyphiidae: Eridantes erigonoides (Emer- ton 1882): This species is common in the no- tillage fields at Horseshoe Bend, where 31 of the 38 adults were captured (Table 4, Fig. 6). It has previously been collected in several states north of Georgia, including Maryland, Tennessee, Virginia, and the District of Co- lumbia (Roth et al. 1988). It is absent from Berry’s (1966, 1970) list. The Horseshoe Bend records are the southernmost known, ex- cept that a male and female were trapped fur- ther south on the upper coastal plain (South Carolina: Barnwell County; Savannah River Site. Pipeline cut with brambles and Prunus; Sifting litter, 28 October 1994. Coll./Det. M. Draney). Linyphiidae: Floricomus tallulae Chamber- lin & Ivie 1944: Two females were trapped in February and seven males in April 1991 in the 342 THE JOURNAL OF ARACHNOLOGY Table 3. — Extended. Species Families Significantly different from present study? 112 25 217 29 No 128 24 Yes 64 17 Yes 45 16 Yes 77 13 Yes 55 13 Yes 22 5 Yes forest. Chamberlin & Ivie (1944) collected this species (then new to science) from Ha- bersham, Hall, and Rabun Counties, Georgia, with the southernmost collection locality be- ing Gainesville (about 40 km north of Horse- shoe Bend). The species is absent from Ber- ry’s (1966, 1970) list, and seems not to have been collected since its description. Linyphiidae: Grammonota inomata Emer- ton 1882: The species is quite common at Horseshoe Bend, where it thrives in the no- tillage fields (Table 4, Fig. 9). The species is known from states north of Georgia, including North Carolina, Tennessee, and Virginia. The records at this site confirm that it thrives in Georgia, but the southern range extension is provided by a male found in the UGA Natural History Museum from the outer coastal plain (Georgia: Tift County; Tifton; Oatfield sweep, December 1963-January 1964. Coll. R. Da- vis, Det. H.E. Frizzell, examined). Linyphiidae: Walckenaeria Carolina Mil- lidge 1983: A single male was trapped in con- ventional tillage winter rye/crimson clover in January 1991. Prior to my finds, it was known from only a few localities in Missouri, North Carolina and West Virginia. This species was described recently (Millidge 1983, holotype collected by J. Berry at Durham, North Car- olina), so range extensions are not surprising. The species appears to be common on the in- ner coastal plain; over 60 individuals of this species were trapped in various habitats in Ai- ken, Barnwell, and Allendale Counties, South Carolina during 13 December 1995-21 Feb- ruary 1996 (Coll./Det. M. Draney). Theridiidae: Paratheridula perniciosa (Keyserling 1886): Several specimens of both sexes were taken in June, July and August 1991 in the conventional tillage field {n = 4) Table 4. — Habitat selection of 15 abundant pitfall species. Table includes all taxa in which at least 20 adults were trapped in 7 monthly 24-hour trap periods of 20 traps/habitat (total = 560 trap/days). Taxa are listed in descending order of number of adults trapped. Habitat abbreviations: C = Conventional tillage field; N = No-tillage field; G = Grassy field borders; F = Floodplain forest. Significance levels: * P < 0.05; ** - p < 0.01; *** = P < 0.0001. ns = Not significant at alpha = 0.05. 2-Way ANOVA Adults in each habitat Habitat 1-way M H Species Total ■ adults C N G F ANOVA means separation Habitat Month inter- action Erigone autumnalis 212 61 48 103 0 G > C, N > F ** ns Pardosa atlantica 175 78 67 30 0 C > G, F; N > F sSssjssJs *** Glenognatha foxi 137 65 66 6 0 C, N > G, F *3!«S(S 5i!5!;5ls Grammonota inomata 104 11 84 6 3 N > C, G, F 5}:** ** Pardosa milvina 55 38 5 12 0 C > N, G, F sfssfssls *** Idionella sclerata 46 0 2 40 4 G > C, N, F ** ns ns Eridantes erigonoides 38 1 31 6 0 N > C, G, F ns ns Phrurotimpus alarms 34 0 3 0 31 F > C, N, G sfs*5{: Allocosa funerea 31 3 8 20 0 G > C, N, F; N > F *** * Ceraticelus emertoni 30 4 1 23 2 G > C, N, F Sj!** ns * Eperigone fradeorum 30 15 1 12 2 C > N, F; G > N sf:* ns Hogna timuqua 26 5 15 5 1 N > C, G, F ** 5js5j::js ns Neoantistea agilis 26 0 3 8 15 F > C, N ** 5jS5)5:i5 ns Tennesseelum formicum 23 12 2 6 3 C > N, F ** ns Pirata iviei 20 2 5 11 2 No differences ns * ns DRANEY— PIEDMONT FLOODPLAIN SPIDERS Pardosa atlantica 343 Pardosa milvina 2 Month Allocosa funerea Hogna timuqua Figures 1-4. — Phenograms of four spider species, family Lycosidae. Graphs illustrate numbers of each stage trapped in 40 traps (10 in each of four habitats) during each of 12 monthly 24-hour trapping periods. Closed circles (•) = males; open circles (o) = females; triangles (A) = immatures; “p” = penultimate instar males; “e” = egg sac; “i” = immatures on female. and the grassy field borders (n ^ 1). This spe- cies is most commonly collected on the outer coastal plain of the gulf coast states, and has been found as far north as Tuscaloosa, Ala- bama (Levi 1957, as P. quadrimaculata (Banks)). The Horseshoe Bend records are a northeastern range extension for the species, which was not listed in Berry (1966, 1970). Besides the three species noted above, sev- eral other Horseshoe Bend species were also missing from Berry’s (1966) list, including Castianeira gertschi, Neriene redacta, Cera- ticelus emertoni, Eperigone inomata, and Idi- onella sclerata. Of these, C gertschi and /. sclerata are recorded from North Carolina and C. emertoni probably occurs there, having been recorded from Virginia (Reiskind 1970; Roth et al. 1988). The remaining five species not yet recorded from North Carolina repre- sent less than 5% of the Horseshoe Bend pit- fall fauna (Table 1), indicating the high degree of similarity of the Piedmont fauna of North Carolina and Georgia. Two of these species, C. emertoni and N. redacta, are also new re- cords for the state of Georgia, although they have been collected in other southeastern states (Roth et al. 1988). Two female N. re- dacta were also collected at the site by J.I. Richardson on 5 September 1967. Phenology. — Twelve spider species were trapped in large enough numbers to give some insight into their life cycles. Phenograms for these species are given in Figs. 1-12. Because pitfall catches reflect the level of activity of a population in addition to its density (Uetz & Unzicker 1976), the numbers trapped should 344 Erigone autumnalis THE JOURNAL OF ARACHNOLOGY Eridantes engonoides Idionella sclerata Month Ceraticelus emertoni Figures 5-8. — Phenograms of four spider species, family Linyphiidae. Graphs illustrate numbers of each stage trapped in 40 traps (10 in each of four habitats) during each of 12 monthly 24-hour trapping periods. Closed circles (•) = males; open circles (o) = females; triangles (A) = immatures; “p” = penultimate instar males. not be interpreted as directly reflecting popu- lation density during the trapping period. In general, male spiders are most active when searching for mates and female spiders are most active when foraging or searching for oviposition sites just prior to and during the period of egg production. Thus the peaks in pitfall catch can be roughly equated with the periods of copulation and egg production (DeKeer & Maelfait 1987). Assuming that im- mature spiders are often food limited (Wise 1993) and likely to be actively foraging, pit- fall catch can be roughly equated with density of immatures. This is likely to be valid only during the warm season, when the immatures are not in diapause or otherwise inactive. Phenograms for four abundant lycosid spe- cies are given in Figs. 1-4. Note the peak in male abundance in March for both Pardosa species, P. atlantica and P. milvina (Figs. 1,2). Both species (the two most abundantly trapped lycosids) seem to have an identical “mating season” after which low numbers of adults continue to be captured until August or September and immatures are trapped until October or November. Berry (1971) docu- ments the seasonal distribution of another Pardosa species, P. parvula Banks (as P. sax- atilis (Hentz)). His pitfall catches show no adults present in March, followed by two peaks of adults. The first peak, in April, was dominated by males (28:1), but by the second, July, peak almost equal numbers of males and females were captured (76 and 65, respective- ly; J. Berry unpubl. data). This could indicate a spring “mating season” peak similar to the DRANEY— PIEDMONT FLOODPLAIN SPIDERS 345 Grammonota inornata Phrurotimpus alarius Neoantistea agilis 1 2 H/lonth Figures 9-12. — Phenograms of four spider species: 9, LinypMidae; 10, Tetragnathidae; 11, Liocranidae; 12, Hahniidae. Graphs illustrate numbers of each stage trapped in 40 traps (10 in each of four habitats) during each of 12 monthly 24=hour trapping periods. Closed circles (•) = males; open circles (o) = females; triangles (A) = immatures; “p” = penultimate instar males. In Fig. 10, squares (■) represent females and immatures of Glenognatha foxi, which could not be reliably separated. Horseshoe Bend Pardosa species, but delayed due to cooler weather than occurred at Horse- shoe Bend in 1991 (a warmer than normal year). Berry (1971) states that the weather during his April collecting period was ‘Very cold and wet.” Alternatively, the later peak of P. parvula could be due to life history prop- erties intrinsic to the species. Allocosa funerea showed a similar, but less pronounced spring peak of males, but in May instead of March (Fig. 3). A high proportion of A. fiinerea individuals was trapped as im- matures, beginning in June. It is unclear whether this indicates that A. funerea adults had a particularly successful reproductive sea- son relative to the Pardosa species, or wheth- er juveniles of this species are relatively more active or the adults relatively less active than is the case with the other lycosid species. These three species probably overwinter as large juveniles, then mature and mate in the spring. This pattern seems to be the rule among smaller lycosids in temperature regions (Doane & Dondale 1979). The larger lycosid Hogna timuqua showed a distinct peak of males in May (Fig. 4). This species probably mates in the spring but most likely needs two years to mature instead of one, perhaps overwintering the second winter as adults, as is the case with other large tem- perate lycosids (Dondale 1977). Immatures of two distinct size classes can be found in the summer (Draney pers. obs.). Adults of many species of erigonine Liny- phiidae were trapped during all seasons of the year (Figs. 5-8). Males of the most commonly 346 THE JOURNAL OF ARACHNOLOGY trapped spider species, Erigone autumnalis were present in all months except January; the catch peaked in May. The few females and identifiable immatures of this species were trapped in June (Fig. 5). The very skewed sex ratio in the pitfall catch probably indicates that the females are not normally very active on the ground surface. Other erigonines which were trapped during most of the year include Mionella sclerata (Fig. 7), Eridantes erigo- noides (Fig. 6), and Ceraticelus emertoni (Fig. 8). The mainly year-round presence of adults and the erratic occurence of both males and females (no clearly defined peak indicating a “mating season”) suggests that these species are probably multivoltine with overlapping generations. This may be the case with E. au- tumnalis as well. Other erigonine species from the southeast are capable of completing their life cycle (egg to egg) in under four months in the lab (Draney unpubl. data), so more than one generation per year is a possibility for these species. One erigonine which seems to display an annual life cycle at Horseshoe Bend is Grammonota inornata (Fig. 9). The sequential appearance of males in January, fe- males in May, and identifiable immatures in June, and the absence of any identifiable in- dividuals in the autumn suggests a strong sea- sonal cycle for this species. The tetragnathid Glenognatha foxi was trapped in all months of the year, with catch increasing from a low in November to a peak in June (Fig. 10). This pattern probably indi- cates an annual reproductive cycle with mat- ing in the early summer. Alternatively, these small tetragnathids may reproduce throughout the year, with population levels and/or dis- persal behavior (thus trap vulnerability) high- est in the early summer. Knowing when im- mature G. foxi exist at the site should resolve the question, but I was unable to confidently separate females from immatures in this spe- cies, so females and immatures are lumped in the data. Berry (1971) graphed the seasonal distribution of G. foxi (as Mimognatha foxi) in his pitfalls. His data show low numbers (< 5 individuals/sample date) of adults (males and females pooled) trapped throughout the year, and low numbers of immatures trapped during June through September. The two sets of data together indicate an annual reproduc- tive cycle for this species in the Piedmont re- gion. Most males of the liocranid Phrurotimpus alarms were trapped in April, with immatures and females found at low levels from Febru- ary to September (Fig. 11). If the hypothesis of an annual cycle with spring mating is cor- rect, then I expect that immatures captured in late winter/early spring would be large sub- adult specimens, whereas those captured in the summer would be smaller immatures that were produced after the spring mating. Peck & Whitcomb (1^8) present pitfall catch data for R alarms which also support the hypothesis of an annual cycle with spring mating. Their catch of males and females (im- matures were not recorded) peaked in May in- stead of April, which may be consistent with the more northerly Arkansas study site result- ing in a delayed “mating season.” At the Ar- kansas site, males disappeared after May, whereas females were trapped through Sep- tember. This pattern is similar to that shown at Horseshoe Bend, again corroborating my life history hypothesis. A similar pattern in the hahniid Neoantistea agilis is also interpreted as an annual cycle with spring mating (Fig. 12). Males peaked in March and females were trapped from March to May. No immatures were trapped, suggest- ing that they spend their time within the leaf litter rather than walking on the ground sur- face. It would be interesting to know whether the few males that were trapped in August- No vember are old adults that survived to au- tumn or whether they are newly adult individ- uals that overwinter as adults. The absence of males in May, June, and July indicate that the latter hypothesis is more likely, N. agilis in Manitoba, Canada apparently displays a dif- ferent life history, with male pitfall peaks in September (Aitchison 1984). Possibly the Manitoba populations are also annual, but a longer period is required for maturation in the cooler cfimate, so the mating is delayed until Autumn. Opell & Beatty (1976) suggest that the species is annual but has two periods of reproduction, in late March to late May and again in mid-August to mid- September. The species may facultatively reproduce during spring and/or autunm, with climate and other conditions determining the local life history pattern. Habitat selection.— The four habitats sam- pled at Horseshoe Bend are all within 10 m of one another and adjacent to one another. DRANEY— PffiDMONT FLOODPLAIN SPIDERS 347 except that the grassy field border separates the agricultural from the forest habitats by a few meters. Since all habitats should be easily accessible to all spider species at the site, sam- pling within the small scale of this agroeco- system landscape enables a determination of the aggregate “habitat preferences” of the spi- der populations. This requires the assumption that numbers of individuals trapped broadly corresponds to the density of individuals in that habitat. Care must be taken since pitfall trapping data has often been shown to violate this assumption (Uetz & Unzicker 1976; Cur- tis 1980; Merrett & Snazell 1983; Topping 1993). Most potential sources of bias have been controlled for in this study. Temporal sources of bias were controlled by trapping simultaneously in all habitats for 24 hours at a time. Effects of weather on spider mobility were controlled for by simultaneous trapping at adjacent sites exposed to identical weather conditions. Interspecific variation in trap vul- nerability is not relevant in this context be- cause abundance comparisons are only made intraspecifically. I acknowledge that habitat architecture can influence the efficiency of the traps (Topping 1993), and that this factor was not controlled for in this study. Even if habitat architecture does affect pitfall catch, species habitat preference should still be identifiable unless architecture has an overwhelming ef- fect on the trappability of species. Comparing catches of species with presumably similar lo- comotory abilities suggests that this is not the case. For example, Grammonota inornata was abundantly trapped in the no-tillage field and rarely trapped in the grass borders (84 and 6 adults, respectively) whereas another erigoni- ne linyphiid, Idionella sclerata, showed the opposite pattern (2 and 40 individuals. Table 4). Such examples of independent catches of apparently similar species in different habitats suggests that architecture is at least not the overriding factor determining pitfall catch, and that habitat preference of individual spe- cies can be examined despite this potentially confounding variable. Some trends in habitat selection are sug- gested by examining the habitats in which the 46 clearly identifiable species in the February- August 1991 data set were trapped. Data for the 15 most abundant of these species are shown in Table 4. One immediately apparent feature is that few species were entirely re- stricted to a single habitat. Although 37% of the species (n = 17) were found in only one habitat, only one species, Floricomus tallulae (not in Table 4; forest, n = 8) was represented by more than three individuals. Another interesting feature of the habitat use list concerns those species which were trapped in all habitats except one. Of the 46 species, 24% {n = 11) were present in three of the four habitats. Eight of these species avoided the forest: Eridantes erigonoides, Er- igone autumnalis, Florinda coccinea (not in Table 4; n = 6), Walckenaeria spiralis (not in Table 4; n = 7), Allocosa funerea, Pardosa atlantica, Pardosa milvina, and Glenognatha foxi. The three remaining species all avoided the conventional tillage habitat: Neoantistea agilis, Idionella sclerata, and Xysticus trigut- tatus (not in Table 4; n = 6). These results are consistent with my expectation that the only habitats that are “avoided” by species with otherwise general habitat requirements are the habitats at either end of a gradient from fre- quently and intensely disturbed and managed (conventional tillage field) to infrequently dis- turbed (forest). Species are less likely to avoid the intermediate no-tillage field and grassy field border habitats. Six species (all linyphiids or lycosids) were trapped in all four habitats: Ceraticelus emer- toni, Eperigone fradeorum, Grammonota in- ornata, Tennesseelum formicum, Hogna ti- muqua, and Pirata iviei. This represents 25% of the 24 species represented by at least four individuals in the data set (and thus theoreti- cally capable of being found in all habitats given their level of activity density). Interest- ingly, four of the five most abundantly trapped species in Table 4 (Erigone autumnalis, Par- dosa atlantica, Glenognatha foxi, and Par- dosa milvina) did not occur in all four habi- tats. This is perhaps contrary to Abraham’s (1983) assertion that dominant spider species in ecosystems tend to be habitat generalists. At Horseshoe Bend, some of the most abun- dant species are habitat specialists at least to the extent that they do not occur abundantly in the forest habitat. Table 4 also presents results of statistical analyses of the habitat distribution of the 15 most abundantly trapped species {n > 19). Two-way ANOVA’s showed that all species except Pirata iviei displayed a significant hab- itat effect. Pirita iviei appears to range widely 348 THE JOURNAL OF ARACHNOLOGY in the floodplain habitats, but was not caught in numbers sufficient to show a significant habitat preference. About half of the remain- ing species showed a significant “month X habitat” interaction, implying that the habitat “preference” of the species changed over time. In some cases, this may be a statistical artifact resulting from low capture rate during certain months. One-way ANOVA’s were performed on all data showing a habitat effect when blocked by month in the two-way ANOVA. Means sep- aration by LSD indicated in which habitats the species were trapped significantly more or less often. Most of these abundant species were much more common in one or two habitats than in the remainder, in which they were rare- ly or never trapped. This pattern of habitat specialization was also observed in Maelfait & De Keer’s (1990) study of spiders in pas- tures and their border zones. Forest species: Only 2 of the 15 abundant species preferred the forest habitat, Neoantis- tea agilis and Phrurotimpus alarms. N. agilis was rarely trapped in either agricultural hab- itat and seems to avoid them. Its prevalence in the forest is consistent with previous col- lection data (Opell & Beatty 1976). Field border species: The thin grassy field borders seem, at first glance, much less a dis- tinct “habitat” than the fields and forest. However, grasslands in Georgia (mostly small strips and patches like the ones in this study) account for about 14% of the land in the state, and can serve as important reservoirs for both beneficial and destructive insects (Morrill 1978). Four species were characteristic of the grassy field borders: the lycosid Allocosa fu- nerea and the linyphiids Ceraticelus emertoni, Erigone autumnalis, and Idionella sclerata. Erigone autumnalis definitely avoids the for- est; none of the 212 individuals were trapped there. Another linyphiid, Eperigone fradeo- rum, was also trapped in considerable num- bers in the grassy borders, though it showed a stronger affinity for the conventional-tillage agroecosystem. Allocosa funerea was also of- ten trapped in the no-tillage agricultural field. This species has often been collected in grassy fields, meadows, and lawns, in addition to gar- dens and pine forests (Dondale & Redner 1983). Duelli (1990) found few species which pre- ferred the grassy margins between cultivated fields and semi-natural (grassland/pasture) ar- eas, and considered the grassy margins in his system to be ecotones, mainly important for harboring species more common elsewhere. However, I have documented several common spiders that were trapped predominantly in the meadow habitats at Horseshoe Bend, indicat- ing that this is their primary habitat, and does not serve as ecotone for them. The grassy hab- itats at Horseshoe Bend undoubtedly also serve as secondary habitats for species which are more abundantly trapped in cultivated fields or forest. In particular, the grassy mar- gins may provide a refuge for cultivated field populations during times when that habitat is disturbed by management practices. No-tillage field species: Hogna timuqua was trapped in the no-tillage agricultural field significantly more often than in other habitats. Two other lycosids commonly trapped here were Pardosa atlantica and Allocosa funerea, though both had stronger affinities to other habitats. Two linyphiids, Eridantes erigono- ides and Grammonota inornata, showed strong preferences for the no-tillage habitat over the other three habitats. I hypothesize that these species must thrive in the thick her- baceous “straw-like” litter layer that is unique to this habitat. Erigone autumnalis was also trapped abundantly in this habitat and in the conventional-tillage field. It seems at first surprising that species would “prefer” the no-tillage habitat to the extent of being much more rarely trapped in both the conventional tillage and the meadow habitats. However, Mangan & Byers (1989) showed that many old-field species remain during the establishment of no-tillage agroe- cosystems from old field habitats. Possibly the “no-tillage” species are adapted to life in ear- ly successional habitats. It seems that the no- tillage habitat may be to some extent ecolog- ically similar to an old-field system for many of these species. Conventional- tillage field species: The te- tragnathid Glenognatha foxi was trapped abundantly in both agricultural habitats but rarely caught in the other two habitats. It is the only abundant species which showed no “preference” for either of the two agricultural habitats. The horizontal orb webs spun by this species were found from about 0.5-3. 0 cm above the ground or litter surface of both hab- itats (Draney pers. obs.); presumably it is de- DRANEY— PffiDMONT FLOODPLAIN SPIDERS 349 pendent on habitat characteristics other than the ground surface architecture. The species has been found in a variety of mostly open, generally xeric situations (Levi 1980), includ- ing meadows, old field, saltmarsh, short grass, and cornfields, which are quite similar to sor- ghum fields. Their considerable ballooning ability (Crosby & Bishop 1936, as Mimog- natha foxi McCook) makes them potentially beneficial colonizers of agricultural fields. Besides G. foxi and Erigone autumnalis, four species were also characteristic of the conventional tillage agricultural field: Pardo- sa atlantica, P. milvina, and the linyphiids Tennesseelum formicum and Eperigone fra- deorum. P. atlantica was found in lower but considerable numbers in the no-tillage field and even in the grassy field border, whereas P. milvina was more restricted to the conven- tional tillage field. Maintenance of biodiversity in agroeco- systems.—-Although more intensive sampling will undoubedly yield additional species, it is clear that the four-habitat agroecosystem at Horseshoe Bend harbors a high diversity of species, similar in structure to that document- ed across an array of successional habitats elsewhere on the Piedmont Plateau (Berry 1966). Much higher species richness can be maintained in agroecosystems composed of a mosaic of habitats under different manage- ment regimes, as is the case at Horseshoe Bend, than in agroecosystems maintained as conventional monoculturai landscapes. This is corroborated by the fact that Bailey & Chada (1968) trapped only 64 species from pitfalls in grain sorghum fields, compared with 112 species I trapped in the more complex sor- ghum/meadow/forest landscape at Horseshoe Bend. Habitat use patterns of individual spider species illustrate two mechanisms which may explain how landscape complexity results in higher spider diversity. First, many species seem to “specialize” in one or a few habitat types; populations may not be able to persist without these habitats. Thus, increasing the number of different habitat types will obvi- ously increase the site-wide richness (gamma diversity) of the agroecosystem as a whole. Second, individuals are often found in habitats other than those in which the species is most abundant. Presumably these species often sim- ply “spill over” to adjacent habitats during foraging and mate- searching behavior from habitats where they are common. This results in higher species richness in each individual habitat (higher alpha diversity) via “mass ef- fect” (Shmida & Wilson 1985). Diffusion of species into suboptimal habitats means these habitats may sometimes serve as refugia {sen- su Duelli 1980), allowing species to persist in an ecosystem when their optimal habitat is disturbed by management practices. One prac- tical effect of this is that species utilizing re- fugia may more rapidly recolonize their pri- mary habitats after the disturbance (plowing, spraying, harvesting, etc.) subsides than would be the case if recolonization were sole- ly by long-distance ballooning. ACKNOWLEDGMENTS I gratefully acknowledge the following for the help they provided in the completion of this project: G.B. Edwards determined the Salticidae; A.J. Penniman determined the Phrurolithinae; G. Hormiga determined Flor- icomus tallulae. D.C. Coleman, D.A. Crossley, E. Gaiser, M.H. Greenstone, R. Han- sen, RF. Hendrix, S. Koponen, and V.L. Med- land reviewed earlier drafts of this manuscript. V.L. Medland provided help with statistical analysis. Finally, I thank J.W. Berry for access to unpublished data. This research was sup- ported by NSF grant BSR-88 18302 to PE Hendrix, and by Financial Assistance Award Number DE-FC09-96SR18546 from the U.S. Department of Energy to the University of Georgia Research Foundation. LITERATURE CITED Abraham, B.J. 1983. Spatial and temporal patterns in a sagebrush steppe spider community (Arach- nida, Araneae). J. ArachnoL, 11:31-50. Aitchison, C.W. 1984. The phenology of winter- active spiders. J. Arachnol., 12:249-271. Anonymous. 1986. (Bathyphantes sp. distribution record). Florida Bureau of Entomology. Tri-olo- gy, 25:3-4. Bailey, C.L. & H.L. Chada. 1968. Spider popula- tions in grain sorghums. Ann. Entomol. Soc. America, 61:567-571. Barnes, R.D. & B.M. Barnes. 1955. The spider population of the abstract broomsedge commu- nity of the southeastern Piedmont. Ecology, 36: 658-666. Berry, J.W. 1966. The distributional ecology of spiders in the old-field succession of the Pied- mont region of North Carolina. Ph.D. disserta- tion, Duke University. 259 pp. Berry, J.W. 1970. Spiders of the North Carolina 350 THE JOURNAL OF ARACHNOLOGY Piedmont old-field communities. J. Elisha Mitch- ell Sci. Soc., 86:97-105. Berry, J.W. 1971. Seasonal distribution of common spiders in the North Carolina Piedmont. Ameri- can Midland Nat., 85:526-531. Blumberg, A.Y. 1979. Comparison of soil surface arthropod populations in conventional tillage, no- tillage, and old field systems. M.S. thesis, Univ. of Georgia. 98 pp. Blumberg, A.Y. & D.A. Crossley, Jr. 1983. Com- parison of soil surface arthropod populations in conventional tillage, no-tillage and old field sys- tems. Agro-Ecosystems, 8:247-253. Chamberlin, R.V. & W. Ivie. 1941. North Ameri- can Agelenidae of the genera Agelenopsis, Cal- lilena, Ritalena and Tortolena. Ann. Entomol. Soc. America, 34:585-628. Chamberlin, R.V. & W. Ivie. 1944. Spiders of the Georgia region of North America. Bull. Univ. Utah (Biol. Series), 35:1-255. Costa, J.T., III & D.A. Crossley, Jr. 1991. Diel pat- terns of arthropods associated with three tree species. Env. Entomol., 20:1542-1548. Coyle, A.E, M.H. Greenstone, A.-L. Hultsch & C.E. Morgan. 1985. Ballooning mygalomorphs: estimates of the masses of Sphodros and Um- midia ballooners (Araneae: Atypidae, Ctenizi- dae). J. Arachnol., 13:291-296. Crosby, C.R. & S.C. Bishop. 1936. Aeronautic spi- ders with a description of a new species. J. New York Entomol. Soc., 44:43-49. Curtis, D.J. 1980. Pitfalls in spider community studies (Arachnida, Araneae). J. Arachnol., 8: 271-280. Doane, J.E & C.D. Dondale. 1979. Seasonal cap- tures of spiders (Araneae) in a wheat field and its grassy borders in central Saskatchewan. Ca- nadian Entomol., Ill :377-380. Dondale, C.D. 1977. Life histories and distribution patterns of hunting spiders (Araneida) in an On- tario meadow. J. Arachnol., 4:73-93. Dondale, C.D. & J.H. Redner. 1983. The wolf spi- der genus Allocosa in North and Central America (Araneae: Lycosidae). Canadian Entomol., 115: 933-964. DeKeer, R. & J.-P. Maelfait. 1987. Life history of Oedothorax fuscus (Blackwall, 1834) (Araneae, Linyphiidae) in a heavily grazed pasture. Rev. Ecol. BioL Sol., 24:171-185. Draney, M.L. 1992. Biodiversity of ground-layer macroarachnida (Araneae and Opiliones) assem- blages from four Piedmont floodplain habitats. M.S. thesis, Univ. of Georgia. 170 pp. Duelli, P. 1990. Population movements of arthro- pods between natural and cultivated areas. Biol. Conserv., 54:193-203. Foelix, R.E 1982. Biology of Spiders. Harvard Univ. Press, Cambridge, Massachusetts. 306 pp. Gaddy, L.L. & J.C. Morse. 1985. Common spiders of South Carolina, with an annotated checklist. South Carolina Ag. Exp. Sta. Tech. Bull. #1094. 182 pp. Gertsch, W.J. & N.I. Platnick. 1980. A revision of the American spiders of the family Atypidae (Araneae, Mygalomorphae). American Mus. Nov., 2704:1-39. Hendrix, RE, R.W. Parmelee, D.A. Crossley, Jr., D.C. Coleman, E.P. Odum & P.M. Groffman. 1986. Detritus food webs in conventional and no-tillage agroecosystems. Bioscience, 36:374- 380. House, G.J. & R.W. Parmelee. 1985. Comparison of soil arthropods and earthworms from conven- tional and no-tillage agroecosystems. Soil & Till- age Res., 5:351-360. Ivie, W. 1969. North American spiders of the ge- nus Bathyphantes (Araneae, Linyphiidae). Amer- ican Mus. Nov., 2364:1-70. Koponen, S. 1992. Spider fauna (Araneae) of the low arctic Belcher Islands, Hudson Bay. Arctic, 45:358-362. Levi, H.W. 1957. The North American spider gen- era Paratheridula, Tekellina, Pholcomma and Archerius (Araneae: Theridiidae). Trans. Ameri- can Microsc. Soc., 76:105-115. Levi, H.W. 1980. The orb- weaver genus Mecyno- gea, the subfamily Metinae, and the genera Pachygnatha, Glenognatha and Azilia of the sub- family Tetragnathinae North of Mexico (Arane- ae: Araneidae). Bull. Mus. Comp. ZooL, 149:1- 74. Maelfait, J.-P. & R. DeKeer. 1990. The border zone of an intensively grazed pasture as a corridor for spiders Araneae. Biol. Conserv., 54:223-238. Mangan, R.L. & R.A. Byers. 1989. Effects of min- imum-tillage practices on spider activity in old- field swards. Env. Entomol., 18:945-952. Merrett, P. & R. Snazell. 1983. A comparison of pitfall trapping and vacuum sampling for assess- ing spider faunas on heathlands at Ashdown For- est, south-east England. Bull. British Arachnol. Soc., 6:1-13. Morrill, W.L. 1975. Plastic pitfall trap. Env. En- tomol., 4:596. Morrill, W.L. 1978. Georgia grasslands: Reservoirs for beneficial and destructive insects. Georgia Ag. Res., 19:25-28. Muma, M.H. 1973. Comparison of ground surface spiders in four central Florida ecosystems. Flor- ida Entomol., 56:173-196. Muma, M.H. 1975. Two vernal ground- surface arachnid populations in Tularosa Basin, New Mexico. Southwestern Nat., 20:55-67. Muma, M.H. & K.E. Muma. 1949. Studies on a population of prairie spiders. Ecology, 30:485- 503. Opell, B.D. & J.A. Beatty. 1976. The nearctic Hah- DRANEY— PIEDMONT FLOODPLAIN SPIDERS 351 niidae (Arachnida: Araneae), Bull. Mus. Comp. Zool., 147:393-433. Platnick, N.I. 1993. Advances in Spider Taxonomy 1988-1991 (P. Merrett, ed.). New York Entomol. Soc., New York. 846 pp. Roth, V.D., D. Buckle & R.L. Crawford. 1988. Linyphiidae of America North of Mexico. Amer- ican Arachnol. Soc., Gainesville, Florida. 62 pp. Shmida, A. & M.V. Wilson. 1985. Biological de- terminants of species diversity. J. Biogeography, 12:1-20. Turnbull, A.L. 1973. Ecology of the true spiders (Araneomorphae). Ann. Rev. Entomol., 18:305- 348. Turner, M.G. & C.L. Ruscher. 1988. Changes in landscape patterns in Georgia, USA. Landscape EcoL, 1:241-251. Uetz, G.W. & J.D. Unzicker. 1976. Pitfall trapping in ecological studies of wandering spiders. J. Ar- achnol., 3:101-111. Wise, D.H. 1983. Spiders in Ecological Webs. Cambridge Univ. Press. 378 pp. Manuscript received 22 July 1996, accepted 5 March 1997. 1997. The Journal of Arachnology 25:352-360 EFFECTS OF PREY SUPPLEMENTATION ON SURVIVAL AND WEB SITE TENACITY OF ARGIOPE TRIFASCIATA (ARANEAE, ARANEIDAE): A FIELD EXPERIMENT Bonnie Jean McNett and Ann L. Rypstra^: Department of Zoology, Miami University, Oxford, Ohio 45056 and Hamilton, Ohio 45011 USA ABSTRACT. The effects of prey capture on web site tenacity and survivorship of Argiope trifasciata (Araneae, Araneidae) were studied in two old field habitats in southwestern Ohio. Adult females were studied in habitats dominated by grass or thistle plants. In manipulation plots, we added two crickets to the webs of approximately half the spiders. We were able to quantify differences in prey intake using morphological measurements that changed with food consumption. The spiders that did not receive sup- plemental food were similar in size to unmanipulated spiders in other areas that we censused. No differ- ences were observed in survivorship or web site tenacity of spiders in grass vs. thistle habitats. No difference in survivorship was observed between fed spiders and those left to natural prey capture. How- ever, spiders receiving supplemental prey relocated their webs less frequently than those spiders that were unsupplemented. The selection of a site in which to live and forage can be a critical decision for a spider since food intake can have a substantial effect on the spider’s ability to survive, grow, and ultimately reproduce (Riechert & Gillespie 1986; Vollrath 1987). The webs that spiders use as foraging tools are energetically costly (Prestwich 1977; Peakall & Witt 1976), and it is not possible for web-spiders to sample their habitat extensively before settling to forage in a particular place (Janetos 1986; Vollrath 1985, 1987). As a result, the initial selection of a site must be based on habitat features and the appropriateness of web attachment sites (Pasquet 1984; Hodge 1987a; Bradley 1993). Once the initial web is constructed, the spider acquires additional information on prey cap- ture which can influence whether it stays or leaves. A number of studies on a variety of species suggest that web-spiders use recent informa- tion on prey capture in deciding whether to stay or leave a particular site (Turnbull 1964; Janetos 1982; Olive 1982; Pasquet 1984; Voll- rath 1985; Gillespie 1987; Rubenstein 1987; Hodge 1987b; Provencher & Riechert 1991; Bradley 1993). A variety of other factors un- related to prey capture, such as the frequency of web destruction or damage, interactions 'To whom correspondence should be addressed. with conspecifics, the spider’s age, and/or the action of predators, can influence a spider’s decision to leave or remain in a given location (Eberhard 1971; Enders 1975, 1976, 1977; Wise 1975; Pasquet 1984; Spiller 1984; Voll- rath & Houston 1986; Gillespie & Caraco 1987; Craig 1987; Smallwood 1993). Clearly, if a particular population is not food limited, prey capture should not have an effect on web site tenacity (Eberhard 1971; Enders 1976; Wise 1993). However, Olive (1981) argues that the phenologies of orb-weaving spiders, particularly those in the genus Argiope Au- douin 1827, are tied to the seasonality of in- sects in their environment and that they evolved under the constraints of food limita- tion. In enclosure experiments with Argiope trifasciata (ForskM 1775), he found that they abandon sites with lower rates of prey capture and aggregate in areas where he supplied prey at a higher rate (Olive 1982). A field study with A. keyserlingi Karsch 1881, revealed that food supplementation, even over a few days, decreased the tendency of individuals to re- locate their webs (Bradley 1993). However, in experiments with A. aurantia Lucas 1833, prey supplementation had no effect on web site tenacity; and the likelihood of wind dam- age appeared to be more critical to the web relocation decision (Enders 1975, 1976). 352 McNETT & RYPSTRA— PREY SUPPLEMENTATION EFFECTS 353 Olive’s (1981, 1982) results influenced us to further investigate the relationship between prey capture, survivorship, and web site te- nacity in A. trifasciata. Since the abdomen of a spider expands as it feeds (Anderson 1974; Jakob et al. 1996), we were able to quantify differences between spiders that receive sup- plemental prey and those left to natural prey capture without disturbing them on their webs. In this way, we were able to verify that the spiders we fed consumed the prey we provid- ed and experienced a change in their overall body condition in some significant way as a result of prey consumption. We then tested the hypothesis that prey supplementation would increase the survivorship and web site resi- dence time of adult females in two structurally distinct habitats. METHODS Study species. — Argiope trifasciata is a conspicuous orb-weaving spider found in gar- dens, tall weeds, and grasses in the eastern United States (Kaston 1948). Spiders emerge from egg sacs in May and June (Kaston 1948). Females mature in September when they are 15-25 mm in length, lay eggs in October and November, and die with the onset of winter (Scheffer 1905; Tolbert 1976). We selected this species for our investigation of the effects of prey capture on web relocation because: (1) their large size makes them easy to monitor in the field; (2) in 1993, the year before this study was conducted, we found them to be very abundant in old field habitats with den- sities as high as 0.82 spiders per m^ (McNett 1995); and (3) although they rebuild the cap- ture spiral of their web each day, they reuse the framework which attaches the web to the vegetation and, as a result, web relocation is costly in comparison to remaining at the same site (Enders 1976; Olive 1981). Study site.— The study population inhab- ited old fields of the Miami University’s Ecol- ogy Research Center, three miles north of Ox- ford, Butler County, Ohio, USA. Two manipulation plots (25 X 20 m) were estab- lished for prey experiments. One manipulation plot was set up in an area dominated by thistle (Cirsium arvense) and the second in an area dominated by grasses (Elymus sp., Fustuca sp. and Phleum sp.). These two habitat types were those that the spiders preferred in 1993 (McNett 1995). Three 5 X 5 m census plots in thistle and three 5 X 5 m census plots in grass, located at least 100 m away from the manipulation plots, were used as control areas in which the spiders were counted and mea- sured but not fed. Prey availability. — Background prey availability was assessed in both the thistle and grass habitats using sticky traps. Each trap consisted of a 20 X 20 cm sheet of plastic to which a thin layer of Tangle Trap® (Tangle Foot, Grand Rapids, Michigan) was applied. Traps were suspended with string that was tied either to natural vegetation or, when neces- sary, to metal reinforcing rods (3.5 m in height). Trap height was randomly determined within the range of 15-92 cm. These values were selected because they corresponded to the range of heights at which webs were found in 1993 (McNett 1995). Trap orientation was determined by randomly selecting a compass direction. A total of nine 400 cm^ traps were hung in each of the six census plots in the early morning of 6 October 1994 and left for 24 h. The arthropods collected were identified to order, counted and measured to the nearest 0.1 mm. Morphological changes in the labora- tory.— Twenty-two adult female A. trifasciata were collected and allowed to establish webs in acrylic plastic (Plexiglas®) cages measur- ing 45 X 45 X 7.5 cm in the laboratory. The total body length and abdomen width of all the spiders were measured after web construc- tion. We selected these measures because it was possible to take them without disturbing the spider in its web. After measurement, nine spiders were fed one cricket (Acheta domes- tica; approximately 150 mg in weight). All of the spiders were then left for 24 h during which time each individual replaced its cap- ture spiral once. At that time, all of the spiders were measured again to determine if morpho- logical differences as a result of feeding would be detectable. During the course of two years of study of this species we were able to obtain morpho- logical measurements in the laboratory of six females before and after eggsac deposition. Spiders were measured and left for 24 h. At that time the egg sac was noted and the spider remeasured. None of these spiders were fed between measurements. Prey manipulation. — On 26-27 Septem- ber 1994, 80 adult female A. trifasciata were 354 THE JOURNAL OF ARACHNOLOGY collected from areas at least 100 m outside the plots and individually marked on their abdo- mens with non-toxic paint. Spiders were held in the laboratory at 15 °C in vials 1 cm in diameter which were not large enough to al- low web construction and therefore minimized any changes in their condition or hunger level. On 28-29 September 1994, all naturally oc- curring A. trifasciata from each of the two 25 X 20 m manipulation plots were removed. In the early morning of 30 September 1994, we introduced 40 randomly-chosen marked spi- ders to each plot by placing them on vegeta- tion approximately 2 m away from other in- dividuals. The next day, we searched the plots and marked the location of each spider’s web by tying flagging to the vegetation near the web. There was a low establishment rate of marked spiders, so all unmarked spiders that moved into the plots after that date were as- signed to a treatment group and included in further data collection. In comprehensive sur- veys conducted in 1993, we discovered that individuals never moved more than two me- ters (McNett 1995), so we were confident in our ability to follow and monitor web site changes of unmarked spiders that moved into our manipulation plots. Introduced as well as unmarked individuals that established in the manipulation plots, were assigned randomly to one of two treat- ments: one group received supplemental prey and the other group was left to natural prey capture. Supplemented spiders received two adult crickets {Acheata domestica; approxi- mately 300 mg) every other day in addition to the prey they captured naturally. Spiders received supplemental prey until they could no longer be located, at which time they were presumed dead. A total of 26 spiders was monitored in the thistle plot, 12 of which were fed crickets, and 30 spiders were monitored in the grass plot, 14 of which were provided with crickets. Spider location was monitored daily from 1 October until no spiders could be found on 27 October 1994. If a spider was not found where it had been the previous day, the sur- rounding 60 m^ area was visually searched. We were able to identify unmarked individu- als by a combination of web location and ab- dominal patterns. Since we never observed a spider move more than 2 m from a previous web site, this large search area eliminated the likelihood that a spider would be falsely as- sumed dead. If we found the spider, we re- corded its new location but, if we were unable to find it, we assumed it was dead. We measured abdomen width and total body length of all spiders in the manipulation plots to the nearest 0.1 mm on 4, 8 and 16 census days after the prey supplementation commenced. On those same dates, we counted and measured all of the spiders in our six cen- sus plots. Statistical analysis.— The number and size of insects captured on sticky traps in grass and thistle were compared with a one-way ANO- VA. The change in body size of laboratory spiders was compared using the t-test. The number of spiders in grass and thistle census plots over the course of the study were com- pared using a repeated measures ANOVA. We compared the abdomen width and body length of field measured spiders in three treatments (supplemented, unsupplemented and cen- sused) in two habitat types (grass and thistle) using an two-way factorial ANOVA and then differences among the specific treatments were compared using Fisher Pairwise Com- parisons. These three groups were compared 4, 8, and 16 census days after the prey sup- plementation was begun. Fed spiders on day 4 would have received prey twice (a total of four crickets), on day 8 they would have re- ceived prey four times (eight crickets), and on day 16 they would have received prey eight times (16 crickets). In order to determine the impact of supple- mental prey on survivorship, the total number of days over which we were able to locate fed and unfed spiders in the two habitats was compared using a two-way factorial ANOVA. In order to determine the impact of prey sup- plementation on web relocations, we also used the two-way factorial ANOVA to compare the movement frequency of fed and unfed spiders in the grass and thistle habitats. RESULTS Spider abundances.—There was no differ- ence between the number of spiders inhabiting thistle or grass in the census plots (Repeated measure ANOVA, F = 1.2, F = 0.3) (Table 1). Census plots had more spiders than we were able to establish in our manipulation plots {F = 18.96, P = 0.032) (Table 1). Since densities in manipulation plots were low com- McNETT & RYPSTRA— PREY SUPPLEMENTATION EFFECTS 355 Table 1. — Number of spiders (mean ± standard error) per square meter in old field habitats domi- nated by grass or thistle. In a two-way ANOVA, there were no differences between densities in grass and thistle habitats (F = 1.2, F = 0.3) but densities in manipulation plots were significantly lower than densities in census plots (F = 18.9, F = 0.03). Plot type Grass Thistle Census 0.52 ± 0.25 0.39 ± 0.12 Manipulation 0.06 ± 0.02 0.05 ± 0.02 pared to natural densities, we believe we suc- cessfully eliminated density as a potentially confounding factor in our study of the effects of habitat type and prey capture on web re- location. Prey abundance.— Sticky traps in grass captured 60.0 ± 10.1 insects in 24 h which was not significantly different from 80.7 ± 20.2 insects captured by these traps in thistle (One-way ANOVA, F = 0.21, F = 0.4). The mean size of the insects captured by sticky traps in grass (2.39 ± 0.19 mm) was also very similar to the mean size captured in thistle (2.46 ± 0.22) (F = 0.07, F = 0.7). The insect orders Diptera and Hymenoptera made up more than 90% of the captures in both habi- tats. Morphological changes. — In the labora- tory, the consumption of one cricket was enough to increase total body length by 0.42 ±0.16 mm, whereas unfed individuals shrank by 0.21 ± 0.16 mm in 24 h (r = 7.35, F = 0.0005) (Table 2). Likewise, the abdomen width of fed spiders increased by 1.34 ± 0.11 mm while the abdomen width of unfed indi- viduals decreased by 0.22 ± 0.15 mm in 24 h (t - 60.70, F < 0.0001) (Table 2). These differences verify that these measurements are an indicator of recent feeding history and spi- der condition. The deposition of an eggsac re- duced both spider abdomen width (t = 8.2, F = 0.0004) and total body length (t = 7.9; F = 0.0005) in the laboratory (Table 2). In ad- dition, the spider’s abdomen appeared shrunk- en and wrinkled after eggs were laid. Prey manipulation. — Although the spiders increased in size over the course of the ex- periment, there were no significant differences observed between spiders inhabiting grass or thistle in the amount that either total body length (Two-way factorial ANOVA, F = 1.34, F = 0.26) or abdomen width (F = 0.24, F = 0.63) changed during the course of the study (Table 3). Unsupplemented spiders within our manipulation plots were not different in either measure of size from the control spiders in the census plots 4, 8 and 16 days after the food supplementation was begun (Fisher pairwise comparisons, F > 0.05) (Fig. 1). However, spiders that received supplemental prey had wider abdomens than spiders in the other two groups (unsupplemented and censused) on all three dates tested (Fisher pairwise compari- sons, F < 0.05) (Fig. 1). Likewise, spiders receiving additional prey were longer than un- supplemented spiders in manipulation plots on all of those same dates and were longer than undisturbed spiders in the census plots on day eight. (Fisher pairwise comparisons, F < 0.05) (Fig. 1). We were able to find fed spiders for 13.6 ± 1.4 days which was not significantly different from the survival of 11.5 ± 1.4 days we ob- served for unfed individuals (Two-way fac- torial ANOVA, F = 0.99, F = 0.33) (Table 3), We were also able to locate spiders in the thistle 14.6 ±1.5 days and in the grass 10.7 ± 1.2 days, but this difference was not sig- nificant at the 0.05 level (F = 3.68, F = 0.06) (Table 3). Fed individuals remained at web sites an average of 12.5 ±1.4 days which was signif- Table 2. — Measurements (mm) of the total body length and abdomen width of female Argiope trifasciata in the laboratory (mean ± standard error). Fed individuals received one cricket (150 mg) whereas unfed individuals and those that produced eggsacs received no food. First measurement After 24 hours Difference Status n Length Width Length Width Length Width Fed 9 14.1 ± 0.8 7.1 ± 0.5 14.5 ± 0.7 8.5 ± 0.5 +0.4 ± 0.2 +1.3 ± 0.1 Unfed 13 15.2 ± 0.5 7.8 ± 0.4 15.0 ± 0.6 7.6 ± 0.4 -0.2 ± 0.2 -0.2 ± 0.2 Produced eggsac 6 14.1 ± 0.8 7.4 ± 0.4 12.5 ± 0.4 5.8 ±0.4 -1.6 ± 0.2 -1.5 ± 0.2 356 THE JOURNAL OF ARACHNOLOGY Table 3. — Results of prey manipulation experiment in which fed spiders in thistle and grass were compared to spiders left to natural prey capture. Morphological measurements (abdomen width and total length) represent the difference between the fourth day after prey supplementation began and the sixteenth day after supplementation began. Data are expressed as mean ± standard error. Fed Unfed Treatment Vegetation Interaction Change in {n = 10) {n - 10) F = 3.61 F = 0.24 F = 1.04 abdomen width P = 0.08 P = 0.63 P = 0.32 (mm) in 12 days Grass {n = 7) 1.50 ± 0.03 1.15 ± 0.55 Thistle (« = 13) 2.18 ± 0.46 0.94 ± 0.20 Both habitats 1.80 ± 0.30 0.98 ± 0.18 Change in body F = 1.22 F = 1.91 F = 0.08 length (mm) in {n = 10) {n = 10) P = 0.28 P = 0.19 P = 0.78 12 days Grass {n = 7) 0.64 ± 0.73 0.10 ± 0.50 Thistle (« = 13) 2.06 ± 0.54 0.83 ± 0.45 Both habitats 1.19 ± 0.45 0.68 ± 0.38 Total days located F = 0.99 F = 3.68 F = 1.71 {n = 26) {n = 30) P = 0.325 P = 0.061 P = 0.187 Grass {n = 30) 11.0 ± 1.8 6.8 ± 1.2 Thistle {n = 26) 14.3 ± 2.3 10.8 ± 2.2 Both habitats 13.1 ± 1.8 8.6 ± 1.6 Web relocations {n = 26) {n = 30) F = 5.60 F = 0.12 F = 2.98 per spider P = 0.022 P = 0.750 P = 0.09 Grass {n - 30) 0.3 ± 0.1 0.4 ± 0.2 Thistle {n = 26) 0.0 ± 0.0 0.6 ± 0.2 Both habitats 0.2 ± 0.1 0.5 ± 0.1 icantly longer than the unfed spiders which remained only 8.7 ± 0.2 days. This difference verifies that fed spiders had significantly few- er web relocations (Two-way factorial ANO- VA, F = 6.99, P = 0.011) (Table 3). Spiders in the thistle relocated their webs with the same frequency as the spiders located in the grass {F = 0.0001, P = 0.95) (Table 3). DISCUSSION An increase in prey capture by adult female Argiope trifasciata influences the decision to relocate or continue foraging in the same web site. These data are consistent with the results of Olive’s (1982) enclosure experiments in which A. trifasciata individuals tended to leave areas in enclosures where food was not provided and aggregate in regions where food was supplemented. The fact that we were able to quantify an increase in spider condition via morphological measurements verifies that the food we were providing was sufficient to af- fect the spiders and provides a close link to food as the mechanism causing the changes in behavior we observed. The fact that we could take these hunger measurements in the field without disturbing the spider is a desirable feature of this system. Since we were able to demonstrate that there was no impact of hab- itat or manipulation on these measures, only the supplemental prey that we provided can account for the differences we observed. Nu- merous studies have associated prey capture with web site tenacity (Turnbull 1964; Janetos 1982; Olive 1982; Riechert & Gillespie 1986; Gillespie 1987; Vollrath 1987; Rubenstein 1987; Bradley 1993 and references therein) but the quantification of prey capture in the past has always been prey in the web rather than some measure of actual intake by the spi- der as we were able to accomplish. One possible confounding factor that might affect our morphological measurements would be the production of an eggsac which sub- stantially reduces the spider’s abdomen size and changes its appearance. However, we did not observe the same kind of emaciation after egg laying in individuals we were monitoring McNETT & RYPSTRA— PREY SUPPLEMENTATION EFFECTS 357 WIDTH (MM) Figure 1 . — Total body length and abdomen width (mean ± SE) of spiders measured 4, 8 and 16 days after prey supplementation was begun. On Day 4, abdomen width was significantly different among groups (F = 4,43, P = 0.003) whereas total body length was not (F == 2.3, P > 0.05). On Day 8, both abdomen width (F = 3.95, P = 0.007) and total body length (F = 2.6, P = 0.048) were signficantly different among the treatments. On Day 16, abdo- men width was signficantly different (F = 4.25, P = 0.0085) but total body length was not (F = 1.389, F > 0.05). in the field that we saw in laboratory spiders. Since eggsacs are deposited very late in the season and since this species produces only one clutch per year (Tolbert 1976), it is likely that the spiders were dying shortly after the production of their egg sacs in the field, per- haps due to an increase vulnerability to pre- dation or other environmental stressors. In any case, we would predict that the spiders receiv- ing food supplements would be most likely to produce eggs since food intake is positively correlated with egg production in many spe- cies of spiders (Wise 1993 and reference therein). Therefore, if egg sac production were confounding our results, it would have re- duced the likelihood of seeing the significant differences in body size we observed between the food supplemented and unsupplemented spiders in this study. Spiders are frequently categorized as food limited in nature because they can survive long periods of starvation (Anderson 1970, 1974), have low metabolic rates (Anderson 1970; Carrel & Heathcote 1976; Nakamura 1987), and the fact that they tend to aggregate in high prey areas (Olive 1982; Rypstra 1989). It has been suggested that the plasticity of the abdomen in spiders is an adaptation to prey shortages because it enables spiders to con- sume large amounts of prey when it is abun- dant and store it for subsequent lean periods (Wilson 1971; Anderson 1974). Since the ability of a spider’s abdomen to expand with consumption should decrease as it reaches its maximum, the substantial morphological changes we observed suggested that the spi- ders in our population were not close to sati- ation. Likewise the fact that manipulating the prey they consumed altered their web site te- nacity provides further evidence that food is a limiting resource for this web-building spi- der (Wise 1993). Food supplementation had more consistent effects on the spider’s abdomen width than on total body length (Fig. 1). The abdomen is flexible and therefore changes size with feed- ing, whereas the cephalothorax is fixed in size for a given instar. Of our measurements, ab- domen width is a more direct measurement of the changes in condition the spider experi- enced since any abdominal changes reflected in total body length are damped by the ceph- alothorax size, which cannot change. As a re- sult, we saw less consistent differences among 358 THE JOURNAL OF ARACHNOLOGY treatments over the course of the experiment in body length than in abdomen width. In ret- rospect, a more accurate assessment of spider condition would have been obtained if we had taken measurements of the cephalothorax alone or some other body part that we knew did not change with feeding. Then we could have scaled body condition on absolute body size as reconamended by Jakob et al. (1996). Optimality theory predicts that the amount of time an organism remains at a site should be related to some combination of prey cap- ture at that site and their investment in that site (Pyke et al. 1977). If this is true then, in a given habitat, spiders with more energeti- cally costly webs should have longer web res- idence times since it should take them longer to recoup the investment in the web itself (Ja- netos 1986; Riechert & Gillespie 1986). The residence times that we recorded for unsup- plemented A. trifasciata were around 8.5 days which is substantially longer than the time re- ported (3 days) for a wide variety of other orb- weaving spiders (Janetos 1982; Olive 1982; Riechert & Gillespie 1986; Smallwood 1993). Even the linyphiids with semi-permanent webs that Janetos (1982) studied had resi- dence times around 5 days. In contrast, resi- dence times of the linyphiid with a semi-per- manent web, Neriene radiata (Walckenaer 1844), were about 10 days; a value much clos- er to those we observed in A. trifasciata (Mar- tyniuk 1983). Since A. trifasciata has a large web and reuses some portion of the support infrastructure, the construction of an entirely new web in a new location may be more cost- ly than the other orb-weavers investigated. The large body size of this spider at late in- stars prevents from moving by ballooning and it appears to walk awkwardly off of the web. As a result, exploring for new web sites is a risky and energetically costly endeavor for A. trifasciata. When spiders reach high densities then in- teractions with one another can influence web site tenacity (Hoffmaster 1986; Rypstra 1985; Smallwood 1993). It seems unlikely that web take-overs or spider interactions on the webs were factors in this study. In experimental plots the spider density was only 0.05 indi- viduals per m^ in the thistle and 0.06 individ- uals per m^ in the grass and the spacing was fairly uniform across the plots. The low den- sities in these experiments and the fact that we never observed individuals moving more than two m in a web relocation event (McNett 1995), suggest that intraspecific interactions were not very important in our these experi- ments. Additionally, it may be that the low densities with which we were working and the elimination of spider-spider interactions as a disturbance, accounts for the relatively long residence times that we observed compared to other orb- weaving spiders. The size and web relocation behavior of A. trifasciata in the grass and thistle habitats we compared were surprisingly similar. Prey cap- ture of spiders in thistle must have been sim- ilar to that in the grass because we uncovered no morphological differences in the spiders inhabiting the two habitats (Table 2). This re- sult is supported by our captures in insect traps which failed to reveal any differences between these two habitats in prey activity at this time in the season. Since Enders (1975, 1976) related web relocation to destruction by wind, we expected to see more relocation events by spiders living in grass since it offers a less sturdy web support than thistle. Perhaps, at least in the season of this study, wind was not sufficiently damaging to affect the spider’s behavior. Although not significant at the 0.05 level, it is tempting to speculate on the nearly sig- nificant difference in survival between ani- mals in the thistle and those in the grass (P = 0.06, Table 2). Indeed, since we were moni- toring such a short period in the end of the spider’s life, it is surprising that there is any suggestion of a difference by habitat in the timing of their death at the onset of winter. Horton (1980) found that Argiope in North American old field habitats are subject to sub- stantial bird predation and that the zig-zag sta- bilamentum offers them some protection from birds. For those of us who have monitored spiders in thistle habitats, it is not difficult to believe that the irritating leaves of this plant could provide the spiders some protection from a variety of vertebrate predators which may have contributed to the near significant difference in survival we observed. In summary, these data demonstrate that change in prey intake is a major factor influ- encing web site tenacity of these large orb- weaving spiders. The difference in body con- dition between spiders that received supple- mental prey and those that did not was the McNETT & RYPSTRA— PREY SUPPLEMENTATION EFFECTS 359 overriding difference between the spiders studied here even though we also compared spiders in two structurally different old field habitats. The ease with which we could verify changes in body condition make detailed anal- ysis of the impact of food intake on the ecol- ogy and behavior of A. trifasciata in a natural setting possible. ACKNOWLEDGMENTS We especially thank Cameron Eicher for his help and support at critical times during this study. C. Ball, L. Barghusen, C. Brandt, G. Cochran. J. Dobyns, & S. Modica also pro- vided valuable assistance with various aspects of this research project. Early drafts of this paper benefited from input from D. Claussen, R. Lee, O. Loucks, B. Steinly, and S. Mar- shall. We are grateful to R. Schaefer for extra patience, advice, and assistance with the sta- tistics. Funding for this project was provided by Sigma Xi, the Scientific Research Society, the Department of Zoology and the Hamilton Campus of Miami University. Voucher spec- imens are available in the Hefner Zoology Museum, Miami University, Oxford, Ohio 45056 USA. LITERATURE CITED Anderson, J.E 1970. Metabolic rates of spiders. Comp. Biochem. Physiol., 33:51-72. Anderson, J.E 1974. Responses to starvation in the spiders Lycosa lenta Hentz and Filistata hiber- nalis (Hentz). Ecology, 55:576-585. Bradley, R.A. 1993. The influence of prey avail- ability and habitat on activity patterns and abun- dance of Argiope keyserlingi (Araneae: Aranei- dae). J. Arachnol., 21:91-106. Carrel, J.E. & R.D. Heathcote. 1976. Heart rate in spiders: Influence of body size and foraging en- ergetics. Science, 193:148-150. Craig, C.L. 1989. Alternative foraging modes of orb web weaving spiders. Biotropica, 21:257- 264. Eberhard, W.G. 1971. The ecology of the web of Uloborus diversus (Araneae: Uloboridae). Oec- ologia, 6:328-342. Enders, E 1975. Change in web site in Argiope spiders (Araneidae). American Mid. Nat., 94: 484-490. Enders, E 1976. Effects of prey capture, web de- struction, and habitat physiognomy on web-site tenacity of Argiope spiders (Araneidae). J. Ar- achnol., 3:75-82. Enders, E 1977. Web site selection by orb-web spi- ders, particularly Argiope aurantia Lucas. Anim. Behav. 25:694-712. Gillespie, R.J. 1987. The mechanism of habitat se- lection in the long-jawed orb- weaving spider, Te- tragnatha elongata (Araneae, Tetragnathidae). J. Arachnol., 15:81-90. Gillespie, R.J. & T. Caraco. 1987. Risk sensitive foraging strategies of two spider populations. Ecology, 68:887-899. Hodge, M.A. 1987a. Factors influencing web site residence of the orb weaving spider, Micrathena gracilis. Psyche, 94:363 — 371. Hodge, M.A. 1987b. Macrohabitat selection by the orb-weaving spider, Micrathena gracilis. Psyche, 94:347-361. Hoffmaster, D.K. 1986. Aggression in tropical orb- weaving spiders a quest for food? Ethology, 72: 929-945. Horton, C.C. 1980. A defensive function for the stabilimenta of two orb weaving spiders (Ara- neae, Araneidae). Psyche, 87:13 — 20. Jakob, E.J., S.D. Marshall, & G.W. Uetz. 1996. Es- timating fitness: A comparison of body condition indices. Oikos, 77:61 — 67. Janetos, A.C. 1982. Foraging tactics of two guilds of web-spinning spiders. Behav. Ecol. Sociobiol., 10:19-27. Janetos, A.C. 1986. Web-site selection: Are we asking the right questions? Pp. 9-22. In Spiders, Webs, Behavior and Evolution. (WA. Shear, ed.). Stanford Univ. Press, Stanford, California. Kaston, B.J. 1948. Spiders of Connecticut. Bulletin 70, State Geol. Nat. Hist. Surv., Hartford, Con- necticut. Martyniuk, J. 1983. Aspects of habitat choice and fitness in Prolinyphia marginata (Araneae: Lin- yphiidae): Web-site selection, foraging dynam- ics, sperm competition and overwintering surviv- al. Ph.D. dissertation, SUNY, Binghamton, New York. McNett, B.J. 1995. Mechanism of habitat selection in the silver garden spider {Argiope trifasciata). M.S. Thesis, Miami Univ., Oxford, Ohio. Nakamura, K. 1987. Hunger and starvation. Pp. 287-295. In Ecophysiology of Spiders. (W. Nen- twig, ed.). Springer- Verlag, Berlin. Olive, C.W 1981. Optimal phenology and body- size of orb-weaving spiders: Foraging con- straints. Oecologia, 49:83-87. Olive, C.W. 1982. Behavioral response of a sit- and-wait predator to spatial variation in foraging gain. Ecology, 63:912 — 920. Pasquet, A. 1984. Predatory- site selection and ad- aptation of the trap in four species of orb-weav- ing spiders. Biol. Behav., 9:3-19. Peakall, D.B. & P.N. Witt. 1976. The energy bud- get of an orb- weaving spider. Comp. Biochem. Physiol., 54:257-264. Prestwich, K.N. 1977. The energetics of web building in spiders. Comp. Biochem. Physiol., 57:321-326. 360 THE JOURNAL OF ARACHNOLOGY Provencher, L. & S.E. Riechert. 1991. Short-term effects of hunger conditioning on spider behav- ior, predation, and gain of weight, Oikos, 62: 160-166. Pyke, G.H., H.R. Pulliam & E.L. Chamov. 1977. Optimal foraging: A selective review of theory and tests. Q. Rev. Biol., 52:137-154. Riechert, S.E, & R.G. Gillespie. 1986. Habitat choice and utilization in web-building spiders. Pp. 23-48, In Spiders, Webs, Behavior and Evo- lution (W.A. Shear, ed.). Stanford Univ. Press, Stanford, California. Rubenstein, D.I. 1987. Alternative reproductive tactics in the spider Meta segmentata. Behav. Ecol. Sociobiol., 20:229-237. Rypstra, A.L. 1985. Aggregations of Nephila cla- vipes L. (Araneae: Araneidae) in relation to prey availability. J. Arachnol., 13:71-78. Rypstra, A.L. 1989. Foraging success of solitary and aggregated spiders: Insights into flock for- mation. Anim. Behav., 37:274-281. Scheffer, T.H. 1905. The cocooning habits of spi- ders. Kansas Univ. Sci. Bull., 111:83-114. Smallwood, P.D. 1993. Web-site tenure in the long jawed spider: Is it risk sensitive foraging or con- specific interactions? Ecology, 74:1826-1835. Spiller, D.A. 1984. Competition between two spi- der species: Experimental field study. Ecology, 65:909-919. Tolbert, W.W 1976. Population dynamics of the orb-weaving spiders, Argiope trifasciata and Ar- giope aurantia (Araneae: Araneidae). Ph.D. dis- sertation. Univ. Tennessee, Knoxville, USA. Turnbull, A.L. 1964. The search for prey by the web-building spider Achaearanea tepidariorum (C.L. Koch) (Araneae: Theridiidae). Canadian Entomol., 96:568-579. Vollrath, F. 1985. Web spider’s dilemma: A risky move or site dependent growth. Oecologia, 68: 69-72. Vollrath, F. 1987. Growth, foraging and reproduc- tive success. Pp. 331-347, In Ecophysiology of Spiders. (W. Nentwig, ed.). Springer- Verlag, Ber- lin. Vollrath, F. & A.L Houston. 1986. Previous expe- rience and site tenacity in the orb spider, Nephila (Araneae, Araneidae). Oecologia, 70:305-308. Wilson, E.O. 1971. The Insect Societies. Belknap Press, Cambridge, Massachusetts. Wise, D.H. 1975. Food limitation of the spider Lm- yphia marginata: Experimental field studies. Ecology, 56:637-646. Wise, D.H. 1993. Spiders in Ecological Webs. Cambridge Univ. Press, Cambridge, England. Manuscript received 9 August 1996, accepted 1 May 1997. 1997. The Journal of Arachnology 25:361-407 PHYLOGENETIC ANALYSIS OF PISAURINE NURSERY WEB SPIDERS, WITH REVISIONS OF TETRAGON OPHTH ALMA AND PERENETHIS (ARANEAE, LYCOSOIDEA, PISAURIDAE) Petra Sierwald: Center for Evolutionary and Environmental Biology, Field Museum of Natural History, Chicago, Illinois 60605 USA ABSTRACT, Apomorphic characters of the Pisaurinae Simon 1898, here recognized as a monophyletic clade comprising 19 nominal pisaurid genera, are described. The genera Perenethis L. Koch 1878 and Tetragonophthalma Karsch 1878 are revised. Three Asian, two African and one Australian species of the genus Perenethis are recognized. The Asian species of the genus Perenethis comprise Ocyala dentifasciata O. Pickard-Cambridge 1885, Tetragonophthalma fascigera Bosenberg & Strand 1906, and Tetragonoph- thalma sindica Simon 1897. A lectotype is designated for the Australian species Perenethis venusta L. Koch 1878. Perenethis parkinsoni Dahl 1908 is regarded as a subjective junior synonym of P. venusta. The African species of Perenethis are Tetragonophthalma simoni Lessert 1916, and Tetragonophthalma symmetrica Lawrence 1927 with its subjective junior synonyms Perenethis huberti Blandin 1975, Pere- nethis lejeuni Blandin 1975 and Pisaurellus badicus Roewer 1961. A lectotype is here designated for the African species Perenethis simoni. Phalaea vulpina Simon 1898 is the only recognized species of the genus Tetragonophthalma with eight subjective junior synonyms: Tetragonophthalma balsaci Blandin 1976, Phalaea crassa Thorell 1899, Phalaea ferox Pocock 1899, T. guentheri Roewer 1955, T. lecordieri Blandin 1976, T. pellengea Roewer 1955, Phalaea thomensis Simon 1909, and T wittei Roewer 1955. Cispius novus Caporiacco 1941 and Cispius tertali Caporiacco 1941 are both subjective junior synonyms of Cispius aethiopicus Caporiacco 1939, now placed in the genus Charminus (NEW COMBINATION). The genera Charminus Thorell 1899, Cispius Simon 1898, Tetragonophthalma, Afropisaura Blandin 1976, Perenethis, Maypacius Simon 1898, and Polyboea Thorell 1895 form a monophyletic group within the Pisaurinae, here called Perenethis genus group. The copulatory organs of this group are figured in detail, the vulval structures for the first time. Males and females of the poorly-known monotypic genus Polyboea are described, and their copulatory organs are figured for the first time. A cladistic analysis of the Pere- nethis genus group is presented. The Afro-Asian distribution of members of the clade Polyboea and Maypacius and the Perenethis-clade is hypothesized to be the result of independent range extensions during the expansion of suitable habitats between the Miocene and the beginning of the Pleistocene. The nursery-web spiders (Family Pisauri- dae) currently contain 54 nominal genera (Platnick 1993), many of them only poorly known. Members of the family are distributed worldwide, displaying great variations in hab- itus, size and life style. Several genera contain large species (up to 30 mm body length) hunt- ing on the surface of freshwater ponds and streams, (e.g., members of the worldwide ge- nus Dolomedes Latreille 1804 and the Afri- can-Asian genus Thalassius Simon 1885 (see Sierwald 1987)), or hunt in trees like spiders of the African genus Tetragonophthalma. Other genera contain small spiders (body length 3-4 mm) hunting on permanent webs, (e.g., in the American genus Architis Simon 1898 (see Carico 1981)). Spiders of the name- bearing Palearctic genus Pisaura Simon 1885 hunt in the vegetation. A well-known species is Pisaura mirabilis (Clerck 1757), famous for the male’s nuptial gift presented to the female during courtship. Morphological data, es- pecially of copulatory organs, life history data and revisionary work at the alpha-taxonomic level, are lacking for most Pisauridae. For many genera, only very few (type) specimens are cataloged and thus accessible in collec- tions. Most American (Carico 1972-1981) and some African (Blandin 1974a-1979b; Sierwald 1987) pisaurid genera were revised recently. The main systematic problem of this family concerns the delineation of the Pisauridae and the definition of subfamilies. No synapomor- 361 362 THE JOURNAL OF ARACHNOLOGY phies have been recognized to date that would distinguish at least the majority of pisaurid genera as a single clade. The often cited nurs- ery-web appears not to be restricted to pi- saurid genera, but similar (homologous?) webs are constructed by Peucetia Thorell 1869 (Family Oxyopidae; Brady 1964; Zahl 1971, color photo of Peucetia nursery- web) and Ancylometes Bertkau 1880 (see Merrett 1988). The systematic position of the latter is uncertain, as it shares characters with mem- bers of the family Ctenidae (eye pattern, re- duced third claw), a group that is presumably not monophyletic (Griswold 1993; Huber et al. 1993). Eye arrangements have a long tra- dition in identification, separation and delim- itation of spider taxa at and below the family level. However, the general “pisaurid” pattern (recurved posterior eye row wider than ante- rior eye row) occurs also at least in the fam- ilies Trechaleidae, Lycosidae, Psechridae, Ctenidae, Acanthoctenidae, and Senoculidae. Formerly, the pisaurid genera were grouped in the three subfamilies Pisaurinae, Thaumasi- inae and Thalassiinae (Simon 1898a; Roewer 1954). Lehtinen’s (1967) suggested placement of “pisaurid” genera in different families (Dolomedidae = Thaumasiinae and Pisauri- dae = Pisaurinae) and superfamilies (Lyco- soidea and Pisauroidea) was poorly substan- tiated, the argumentation lacking supportive evidence in form of clearly defined synapo- morphies (Brignoli 1983; Sierwald 1990). New catalogs (Platnick 1989, 1993) listed pi- saurid genera without reference to subfami- lies. Recent progress in systematic studies iden- tified 10 genera, originally assigned to Pisaur- inae and Thaumasiinae, as a monophyletic clade. Based in part on characters of the eye arrangement and synapomorphies in male and female copulatory organs (Sierwald 1993), these 10 genera were placed in the re-erected South American family Trechaleidae Simon 1890 (Carico 1986, 1993). The present study describes the defining characters of a mono- phyletic clade consisting of 1 8 pisaurid genera all related to the genus Pisaura and for which Simon’s name Pisaurinae is available (Tables 2, 3). This study also presents a cladistic anal- ysis of a monophyletic clade within the here redefined Pisaurinae, the Perenethis genus group. This group contains the genera Char- minus Thorell 1899, Cispius Simon 1898, Af- ropisaura Blandin 1976, Tetragonophthalma Karsch 1878, Perenethis L. Koch 1878, May- pacius Simon 1898 and Polyboea Thorell 1895. The majority of characters used in the cladistic analysis stem from the copulatory or- gans and the internal female organs for these taxa are figured for the first time. Taxonomic revisions of the genera Tetragonophthalma and Perenethis are included. The distribution of members of the Perenethis genus group is noteworthy and discussed below. Charminus, Cispius, Afropisaura, Tetragonophthalma and Maypacius are restricted to Africa, the mono- typic genus Polyboea to Asia, and the genus Perenethis is widely distributed in Africa, Asia, and Australia. The African-Asian distri- bution pattern observed in this group of taxa occurs in other pisaurid and non-pisaurid spi- der groups as well. The cladogram obtained through the phylogenetic analysis suggests hypotheses regarding the origin of this distri- bution pattern. METHODS Specimens were made available by the in- stitutions and their curators listed in the Ac- knowledgments. The institutional acronyms were taken from Arnett et al. (1993). Speci- mens designated and published as allotypes by original authors are in fact paratypes, speci- mens designated as neallotypes in subsequent publications have no validity under ICZN reg- ulations (ICZN 1985). Phylogenetic analysis of the Perenethis genus group. — Outgroup: The African pi- saurid genera Charminus and Cispius, them- selves sister taxa, serve as out-groups in the cladistic analysis. The membranous, sac-like anterior section of the female copulatory duct as it occurs in Charminus camerunensis Tho- rell 1899 (Fig. 5) and in all genera of the in- group is the synapomorphy for ingroup and outgroup. The sister-group relationship of Charminus and Cispius is in this data set sup- ported by one synapomorphy, the procurved ridge (= carina) of the epigynum (Figs. 4, 6). Characters: Character scoring is presented in Table 4. The character matrix contains 35 characters (17 binary, 18 multistate) with 98 states: 17 characters with 45 states from male copulatory organs, 11 characters with 31 states from female copulatory organs, and 7 somatic characters with 22 states. An artificial amalgam taxon of the genus Cispius, combin- SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 363 ing the female characters from Cispius var- iegatus Simon 1898 (type species of the ge- nus, male unknown) and the male characters of Cispius thorelli Blandin 1978 (female un- known) was used in the data matrix. This pro- cedure became necessary due to the scarcity of males in identified museum collections. The copulatory organs of Cispius maruanus (Roewer 1955), the only species for which both sexes are known, are similar to varie- gatus and thorelli respectively (Blandin 1978a, figs. 8, 19; only a single male speci- men, the paratype, of maruanus is known). Tree generation: Trees were generated with the software package Hennig86, version 1.5 (Farris 1988), using the “ie^” command (im- plicit enumeration, calculates all possible shortest trees). All multistate characters were treated unordered (non- additive, implemented with the “ccode-;” command). Character op- timization: Characters were optimized on the trees using CLADOS, version 1.2 (Nixon 1992), which permits comparing ACCTRAN and DELTRAN optimization (implemented in “Un Equis” Mode, using the commands “o” [ACCTRAN] and “CTL o” [DELTRAN]. Figure 2 shows the ACCTRAN optimization of characters, and Table 5 notes the characters that differ under DELTRAN optimization. Characters were mapped on the cladogram shown in Fig. 2 using the HOM = 0 (default) settings: Only those character state changes are indicated as homoplastic (white rectan- gles) that designate more than one indepen- dent origin of that state within this data set. Non-homoplastic character state changes are indicated as black rectangles. Usage of terms. — Character: The term re- fers to an actual structure (e.g., conductor). Identical terms used for particular structures in male and female copulatory organs imply homology. A particular structure may appear in several different conditions (= character states). Morphologically indistinguishable character states were given identical codes in the data matrix, without a priori regard for the distribution of that particular state within the 14 species. To facilitate precise comparison of positions of various elements in the male gen- ital bulb, the terms “proximal” and “distal” were used as defined below. Male palpal or- gans: Differing from common usage, the terms “proximal” and “distal” on the tegul- um of the male palp refer to the position of a particular part in relation to the trajectory of the sperm duct. In all Pisauridae studied so far, the reservoir of the sperm duct forms a single spiral within the tegulum (Sierwald 1990, fig. 2). Following the course of the sperm duct, starting at the fundus in the sub- tegulum, apophyses inserting on the tegulum near the fundus are considered proximally lo- cated. Apophyses inserting closer to the ejac- ulatory duct, i.e., the embolic base, are re- ferred to as being distal. Often these topological data can only be assessed in the expanded bulb. The terms dorsal, ventral, ba- sal, apical, retrolateral, and prolateral are used descriptively for positions within the unex- panded bulb in ventral view. Dorsal and ven- tral refer to the position of a particular part being within the depth of the unexpanded gen- ital bulb. Ventrally located parts are mostly visible in the unexpanded bulb in ventral view (as figured here). Female copulatory organ: The term epigynum refers to the external parts, the term vulva to the internal parts of the female copulatory organ. Copulatory duct describes that part of the duct that connects the copulatory opening with the spermatheca. Fertilization duct defines that part of the duct that connects the base of the spermatheca with the uterus extemus. The spermatheca in Pi- sauridae consists of a rounded head, bearing pores, attached to the stalk of the spermatheca. The stalk of the spermatheca, containing the spermathecal duct, connects to the base of the spermatheca (Fig. lb). Dissections, measurements, drawings. — Dissections and drawings follow the proce- dure described in Sierwald 1989b, 1990. All measurements for body length, prosoma length and width and leg length are in mm. Recorded leg length in species descriptions is given for leg I. Leg segments were measured dorsally. Relative spine length is compared between taxa. The tibial spines are rather uni- form in length on all legs of an individual and similar in length to most metatarsal and fem- oral spines (spines notated as “1” in Table 6), except when noted otherwise (spines notated as “i” or “I” in Table 6). To assess relative spine length a ratio was calculated by dividing the absolute length of one spine of the second pair of ventral spines of tibia I by the width of tibia I. The calculation was repeated for up to five individuals if available. A ratio of three indicates the spine being 3X as long as the 364 THE JOURNAL OF ARACHNOLOGY Figure 1. — Pisaurine copulatory organs, schematic, la, Female epigynum; lb. Female vulva. Ic, Male palp. tibia is thick. Eye arrangements were recorded in frontal view. The lens was measured in side view, perpendicular to the optical axis. Small- er specimens generally have relatively larger eyes than larger specimens of the same spe- cies. Eye size ratios may be reversed in ex- tremely small or large specimens. Table 1 lists the abbreviations of terms. PHYLOGENY AND ZOOGEOGRAPHY The Pisaurinae Simon 1898. — Nineteen nominal genera listed in Table 2, including the genus Pisaura Simon 1885, form a clearly de- marcated monophyletic group. All of these genera were originally placed in Simon’s Pi- saureae (Simon 1898), except the more re- cently described genera Thalassiopsis Roewer 1955, Euprosthenopsis Blandin 1974 and Vuattouxia Blandin 1979 (Table 2). For this clade, Simon’s subfamilial name Pisaurinae is available. Table 3 lists genera previously as- signed to the Pisaurinae, which are now re- moved from the here newly defined Pisauri- nae. The affinities of the genus Walrencea Blandin 1979 are currently unknown (no spec- imens examined). The non-pisaurine genera listed in Table 3 and the remaining pisaurid genera formerly listed in the Thalassiinae and Thaumasiinae (Roewer 1954) cannot yet be grouped into monophyletic clades. Griswold (1993) noted in his analysis of lycosoid rela- tionships that Dolomedes differs from Pisaura largely through the retention of plesiomorphic characters. Synapomorphies presently known for the Pisaurinae are: 1) The presence of a distal tegular apophysis in the male genital bulb (Fig. Ic) in addition to the median apophysis and the conductor, and 2) the pres- ence of a ridge (= carina) with a pair of pits (= fossae) in the female epigynum. All mem- bers of the Pisaurinae for which the vulval structure is known have relatively small sper- matheca, positioned at the posterior end of the vulva, with the copulatory duct communicat- ing with the stalk of the spermatheca (Figs, la, lb; for comparison with other pisaurid genera see Sierwald 1989b). Characters of the Pisaurinae: Eye pattern: Eyes form two rows, PER recurved and wider than AER; PLE on low tubercles, PME equal or slightly smaller than PLE. AER recurved, straight or procurved. Several distinct char- acter states of strongly procurved AER can be distinguished, i.e., character 2 states 2-4. ALE may be on tubercles in procurved AER. Eye sizes and ratios vary depending on size of specimens (see Methods). Chelicerae: Anteri- or margin always with three teeth, the middle tooth twice as large as the equally- sized lateral teeth. Posterior margin with 2, 3, or 4 teeth (Table 2), often genus specific. Spine pattern (Table 6): Spination of legs rather homoge- neous within the subfamily, limited genus spe- cific variations affecting few spines occur. Color pattern: Basic coloration of cuticle yel- lowish-brown to dark brown; gray to nearly black diffuse coloration located in the tissue directly beneath the integument (sternum, legs); white guanine in the opisthosoma, dor- sally and ventrally. Patterns produced by col- ored hairs, dorsally two yellowish-to-white SffiRWALD— PHYLOGENY OF PISAURINE SPIDERS 365 Table 1. — Abbreviations on figures and in text. Copulatory organs A sclerite A at base of embolic division (c3) bk basal hematodocha ( c? ) bmt basal membranous tube (embolic di- vision, S) bs base of spermatheca ( 9 ) c conductor (d) ca Carina of epigynum ( 9 ) cd copulatory duct (9) CO copulatory opening (9) db dorsal branch of dta (c?) ds spermathecal duct ( 9 ) dst distal sclerotized tube (embolic divi- sion, (3) dta distal tegular apophysis (c3) dtp distal tegular projection (c?) e embolus (d) epf epigynal folds ( 9 ) fd fertilization duct ( 9 ) fo fossae, epigynal pits on carina ( 9 ) hs head of spermatheca ( 9 ) ll lateral lobes of epigynum (9) If lateral flap at conductor ma median apophysis (c?) mf middle field between // ( 9 ) p petiolus (c3) pp pars pendula of embolus ((3) rta retrolateral tibial apophysis ( c? ) ss stalk of the spermatheca ( 9 ) St subtegulum (<3) t tegulum ((5) tr tnincus of embolus (d) vb ventral branch of dtp {S) vta ventral tibial apophysis (c?) Legs Fe femur Pa patella Ti tibia Me metatarsus MeTa metatarsus-tarsus PaTi patella-tibia Eyes AE anterior eyes AER anterior eye row ALE anterior lateral eyes AME anterior median eyes PE posterior eyes PER posterior eye row PLE posterior lateral eyes PME posterior median eyes Miscellaneous ch character, see Table 4 longitudinal stripes, rows of dark brown-to- black hairs often contrasting with the bright stripes. These color pattern elements occur also outside the Pisaurinae within the Pisaur- idae sensu lato, e.g., in Dolomedes. Leg spi- nation (Table 6): Spination pattern uniform on certain leg segments throughout the Pisauridae sensu lato (e.g., Fe II, III), others show sev- eral, often genus-typical character states (e.g., patellar spination; see below in description of genera). Spine-length ratio: Small, thin-leg- ged, web-living Pisauridae (e.g., Polyboea) with longer spines than medium-sized species hunting in vegetation (e.g., Charminus, Pisau- ra); short spines predominantly in large-sized species (e.g., Maypacius and Tetragonoph- thalma) (see Methods for calculation of ratio). Female copulatory organ: (Figs. la,b). As in other Lycosoidea, the female copulatory or- gan consists of two lateral longitudinal folds, the epigynal folds (Sierwald 1989b). The in- ternal pouches of these folds each contribute a copulatory duct, a spermatheca and a fertil- ization duct to the vulva. In the Pisaurinae, the epigynal folds run longitudinally, diverging (e.g., Perenethis) or converging anteriorly (e.g., Charminus) or forming curves (e.g., Pi- saura). The copulatory opening is situated along the anterior section of the folds, thus the trajectory of the anterior section of the epi- gynal folds determines the position of the cop- ulatory openings relative to other features within the female organs and the course of the anterior part of the copulatory duct in the vul- va, e.g., whether it starts medially or laterally. In the genera Afropisaura, Tetragonophthal- ma, Perenethis, Maypacius, and to a lesser de- gree, Polyboea, the epigynal folds diverge strongly anteriorly, thus placing the copula- tory openings laterally. Epigynum: The integ- ument between (middle field, mf) and around the folds (lateral lobes, //) is often strongly sclerotized and may form projections, grooves, pits, hoods, ridges, etc. When com- paring the genera Pisaura and Afropisaura, Blandin (1976b) suggested interesting homol- ogy-hypotheses concerning the transverse ridge and the two pits (Fig. la). These ele- ments can be identified in the epigyna of all 18 pisaurine genera. The integument anterior to the epigynal folds forms a transverse ridge (carina, ca) that can be straight, procurved or recurved, forming a large sclerotized lip, can be entire or separated in two branches. The Table 2. — Variable characters in the Pisaurinae. * Originally described by Simon (1898a: 295) as Caripeta (name preoccupied), Caripetella nom. nov. by Strand (1928). ^ Data from Blandin 1979a. ^ Considered a synonym of Nilus by Simon (1889a: 296). 6 males only known, 9 females only known. Character 366 THE JOURNAL OF ARACHNOLOGY c . o ^ a g o -S ^ 3 u cd , 2 ^ 2 U a. o a S.S ^ (U ic ■> fl - ^ § s ’3 ^ ■> >. % O .5 ^ c« .3 -S 60 d •a s o 0 d .2 d ^ V. 1 I ^ 'd o ^ ^ o d p ^ 60 0) g d 2 ^ 3 .3 M O p p p p p p Pd > CZ) HH HH HH HH T3 -P T3 T3 •p -p -p pj T) d •p •d •p d d d "P d d d T3 d d > d d d > > > d > > > d > > u > > > U tH tH > tH tH > tH P Wm tl P P P tH p d P tH P d p p p d d d P d d d P d d O d d d 0 o O d P 2 p P d P 2 u a 2 2 2 tl Oh tH Oh Oh 2 Oh a Oh 2 tH Oh V V V V V A II II II V V V A V II II ^AHAVVAAAVAVAAAHA K4l-gKqKqi^K^h,4KlKqi^KjKqNqi>qKqsg»^ AAAAAAAAAVAAAAAHA a,a,a,Q,a,Q,a,Q„a,a,a,a,a,ei,a,a,ci, ^ ^ os "d ^ I 2 ^ o m ^ -S S 5^ S pSJ U 00 0\ 00 00 0® O ON 00 ^ Os ^ 60 ll d o a o O 60 S .a w g s "S' :i6 O O* 6^ Q Ci, o C $6 •S 2 § Q OS d ;p 00 d os U 2° S d •is § S § c ^ so r- Os 00 d ■-o 22 d m P 00 ^ U 00 ffl o ^ d S J 21 tl 'S « ^ ^ 5 ? s S ^ S: g »n Os - S Os S 60 2 o d xi o H S § ^ § o d 0, (k; 00 r- 00 o Ifl OS ^ 00 ^ 00 ^ a c J "1 ■I § I a SI o. S Thalassiopsis Roewer 1955 PME>AME ALE>AME ALE=PME recurved 3 equal Vuattouxia Blandin 1979 PME>AME ALE>AME ALE^PME procurved 3 equal SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 367 Table 3. — Systematic position of genera listed previously in the Pisaurinae by Simon (1898:282-297) and Roewer (1954:110-126). * Type species congeneric with Dolomedes based on description in Roewer (1955:190, figs. 73a, b). Also discussed by Blandin (1978:38). ^ Listed in Simon's Supplement Generate (1898:1045) and assigned to the Dolomedeae. ^ Specimens examined for this study. Listed in the Sup- plement Generate (1898:1044) and assigned to Pisaureae. Simon 1898 Roewer 1954 Current systematic position Architis Simon 1898 Architis Pisauridae sensu lato Cispiolus Roewer 1955 synonym of Dolomedes^ Cispiomma Roewer 1955 synonym of Cispius (by Blandin 1978a:44) Enna Cambridge 1897^ transferred to Trechaleidae (by Carico 1986) Euprosthenomma Roewer 1955 synonym of Euprosthenops (by Blandin 1976:67) Eurychoera Thorell 1897 Eurychoera Pisauridae sensu lato^ Ischalea L. Koch 1872 Ischalea transferred to Stiphidiidae Melocosa Geitsch 1937 transferred to Lycosidae Pelopatis Bishop 1924 Synonym of Pisaurina (by Carico 1972:297) Phalaea Simon 1898 synonym of Tetragonophthalma Pisaurina Simon 1898 Pisaurina Pisauridae sensu lato Sisenna Simon 1898 Sisenna transferred to Trechaleidae (by Sierwald 1990:51) Spencerella Pocock 1898^ Spencerella synonym of Chiasmopes (by Blandin 1974a:311) Staberius Simon 1898 Staberius Pisauridae sensu lato Thanatidius Simon 1898 Thanatidius synonym of Pisaurina (by Carico 1972:297) Genera of currently unknown affinities Cispinilus Roewer 1955 no specimens examined Nilus O.P. — Cambridge 1876 Nilus no specimens examined Phalaeops Roewer 1955 no specimens examined StoUczka O.P. — Cambridge 1885 StoUczka no specimens examined Carina possesses two lateral pits here termed fossae (fo). The fossae may be located directly above the copulatory openings, lateral to them or between them. Their positional relationship to the copulatory opening and additional de- pressions (e.g., as in Afropisaurd) probably determine the fixation mechanism needed to securely connect the expanded male bulb and the epigynum. Thus, the Pisaurinae invite the study of copulation mechanics and its evolu- tionary change. Vulva: The slitlike copulatory openings lead into the mostly membranous, saccate copulatory ducts, often forming loops. In the genera Afropisaura, Tetragonophthal- ma, Perenethis, and Polyboea the copulatory duct possesses two large saccate loops, with the first loop running from the lateral copu- latory opening towards the middle of the vul- va. The copulatory ducts {cd) enter the stalk (ss) of the laterally located spermatheca close to the perforated spermathecal heads (hs). The coiled spermathecal ducts (ds) enter the large base of the spermatheca (bs), which contains either an enlarged lumen or additional coils of the duct. The coiling pattern of the sperma- thecal duct is often species-specific. The fer- tilization duct (fd) is short and sclerotized, originating at the medial portion of the sper- mathecal base. Such tri-partite spermatheca, consisting of head, stalk and base, has been found in many genera of Lycosoidea (Jarvi 1905; Sierwald 1989b, Griswold 1993) and other members of the RTA-Clade (which com- prises all families in which males possess a retrolateral tibial apophysis; Coddington & Levi 1991). Male copulatory organ: The male copula- tory organ of Pisaura has been described and figured in detail elsewhere (Sierwald 1990). The most conspicuous feature is the presence 368 THE JOURNAL OF ARACHNOLOGY of three apophyses, the conductor {co) on the retrolateral side, the median apophysis {md) and distal tegular apophysis {dta) ventrally in the center of the unexpanded bulb (Fig. Ic). The distal tegular apophysis represents a con- spicuous synapomorphy for the Pisaurinae. The distal tegular apophysis was labelled ful- crum by Blandin (1976b) but is not homolo- gous with the “fulcrum” sensu Comstock (1910) in Dolomedes (Sierwald 1990). In con- trast to other pisaurid genera, members of the Pisaurinae possess a simple retrolateral tibial apophysis {rtd) with a single rounded or point- ed tip. A petiolus is well developed. The tegu- lum is bowl-like, with the sperm duct follow- ing the outer margin of the bowl. The upper surface of the tegulum is partly sclerotized but membranous around the base of the median apophysis and around the ventral branch of the distal tegular apophysis, permitting both to tilt out of their position during inflation. The distal tegular projection {dtp) is broad- ened and sclerotized, some with one or two humps. The base of the distal sclerotized tube of the embolic division appears to be sup- ported by the sclerotized humps of the distal tegular projection during expansion of the bulb (Fig. 11). The conductor (co), an outgrowth of the re- trolateral tegular wall, displays several genus- typical character states within the Pisaurinae. An often strongly-sclerotized outer retrolateral wall and a more or less membranous inner prolateral wall that is partly inflatable are the basic components. The inner wall may feature membranous folds, along which the embolus rests in the unexpanded bulb (Figs. 8, 10). The conductor is a very important feature for the analysis of the Pisaurinae, since various parts of the conductor display several states (see un- der Character description below). The slender median apophysis {ma) is shorter than the dis- tal tegular apophysis, with a sclerotized point- ed tip often forming a hook. Its basal and pro- lateral sections are membranous and expandable. The large distal tegular apophysis {dta) has a dorsal {db) and a ventral branch {vb). The ventral branch anchors the distal apophysis in the tegulum, the dorsal branch connects to the basal membranous tube {bmt) of the embolic division. The shape of the ven- tral branch resembles a scapula and may carry a wing-shaped flap apically (Figs. 54, 57, 64). The ventral and dorsal branches are joined apically, forming a hook-shaped beak pointing retrolaterally. Next to the dorsal branch, with- in the basal membranous tube of the embolic division lies a sclerite labelled “A” in Figs. 11, 14, 29, 94, 96, 99. The sclerite is present in Pisaura (see Sierwald 1990, fig. 45, not labelled). In most species sclerite A is visible only in expanded palps. The embolic division, connected to the distal tegular projection via a membranous tube (basal membranous tube, bmt), consists of the distal sclerotized tube {dst), fused to the truncus of the embolus, and the pars pendula {pp). The embolus is of vary- ing length and often whiplike. Natural history. — Data on behavior and life history are scarce. Apparently, the major- ity of pisaurine spiders hunt in the vegetation. Members of a few genera, Euprosthenops Po- cock 1897 (see Gerhardt & Kastner 1938) and Polyboea (see Koh 1989) build webs for prey capture. Very young Pisaura mirabilis hunt in webs (Lenler-Eriksen 1969); their webs re- semble those built by young Dolomedes and young Pisaurina Simon 1898, and the per- manent webs built by Architis (see Carico 1985; Nentwig 1985; Sierwald 1990). The copulatory behavior of Pisaura mirabilis is well known: The male presents a wrapped prey item to the female (Hasselt 1884; Bris- towe 1958; Nitzsche 1988). Unfortunately, the copulatory behavior of other pisaurine species is unknown. Nursery-webs have been reported from Pisaura mirabilis, Afropisaura and other pisaurid genera outside the Pisaurinae (Sier- wald 1990). Specimens examined. — Members of the genera Afropisaura, Charminus, Cispius, Maypacius, Per- enethis, Polyboea, and Tetragonophthalma exam- ined for this study are listed below under the de- scription of each genus. Other material: Caripetella madagascariensis (Lenz 1886): MADAGASCAR: Fianarantsoa Province, P.N. Ranomafana, Talatak- ely, 2ri5'S, 47°25'E, 900 m, 19, 5-7 October 1993 (Scharff, Larcher, Griswold, Andriamasiman- ana) (currently CASC). Toamasina Province, P.N. Perinet, 1000 m, near Andasibe, 18°56'S, 48°24'E, 29, 4-5 November 1993 (Coddington, Larcher, Griswold, Andriamasimanana, Scharff) (currently CASC). Antsiranana Province, Marojejy Reserve, 8.4 km NNW Manantenina, 14°25'S, 49°45'E, 700 m, numerous 6 9, 10-16 November 1993 (current- ly CASC). Chiasmopes namaquensis (Roewer 1955): SOUTH AFRICA: Cape Province, Die- Vlug, near Avontuur, fynbos dung trap, 19, 16-19 December 1981 (S. & J. Peck) (AMNH). Chias- SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 369 mopes sp.: SOUTH AFRICA: Natal, Cathedral Peak Forest, 75 km WSW of Estcourt, grassland, pan trap, 2cJ, 13-31 December 1979 (S. & J. Peck) (AMNH). C hystrix (Berland 1922): SOUTH AF- RICA: Transvaal, Ohrigstad, 14 km S Belfast, 1 9, 27-29 December 1990 (V.D. & B. Roth) (CASC). Dendrolycosa sp.: PHILIPPINES: Luzon, Ilocos Norte, Pagudpud, Subec, 1(3, 23 May 1987 (C.K. Starr) (USNM). MALAYSIA: Perak, Cameron Highlands, 19 (Koh 84.06.12.10). Pahang, Fraser’s Hill, 1 9 (Koh 76.1 1.16.08). BRUNEI: 1 9 with egg sac, (Koh 83.01.26.02). PAPUA NEW GUINEA: Madang Province, Nobonob Hill, 7 km NW Ma- dang, 5°10'S, 145°5'E, 2c3, 1 May 1988 (W.J. Pu- laski) (CASC). Euprosthenops australis Simon 1898: TANZANIA: Inside W. slope Ngorongoro Crater, 1850 m, 19, 21 October 1957 (E.S. Ross & R.E. Leech) (CASC). Seronera, Serengeti National Park, 1450 m, 19, 24 November 1967 (E.S. Ross & A.R. Stephen) (CASC). Euprosthenops bayaon- ianus (Capello 1866): ZIMBABWE: Kariba, 26, 16 August 1990 (V.D. & B. Roth) (CASC). ZAIRE: Faradje III. Lessert det., 1 9 (AMNH). Euprosthenops biguttatus Roewer 1955: ZIMBA- BWE: Mazabuka, Matthysse leg., 1(3, 17 Septem- ber 1952 (AMNH). ANGOLA: Lunda Province, Nova Chavez, 19, 14-16 September 1949 (B. Malkin) (CASC). Euprosthenopsis sp.: SOUTH AFRICA: Transvaal, Klaserie, Guernsey Farm, 1 (3, 18-31 December 1985 (S. & J. Peck) (AMNH). Cape Province: Table Mountain, Skeleton Gorge, 34°S, 18°30'E, 19, 13 February 1991 (V.D. & B. Roth) (CASC). KENYA: Rift Valley Province, Lake Naivasha, Fisherman’s Camp, ca. 0°45'S, 36°20'E, 1(3, 19 October 1992 (V.D. & B. Roth) (CASC). KENYA: 49 mi NW of Mobasa, 450 m, 19, 4 November 1957 (E.S. Ross & R.E. Leech) (CASC). TANZANIA: NE side of Mt. Mem, 1500 m, 1 9, 28 October 1957 (E.S. Ross & R.E. Leech) CASC). Euprosthenopsis armatus (Strand 1913): ZAIRE: Garamba, 16 (AMNH). ZIMBABWE: Harare, 3 9, 14 August 1990 (V.D. & B. Roth) (CASC). Eurychoera quadrimaculatus Thorell 1897: SINGAPORE: MacRitchie Reservoir, 19 (Koh 77.01.01.01), 1(3 (Koh 88.020.06). Paracla- dycnis vis Blandin 1979: MADAGASCAR: Anta- nanarivo, 19, several juv., 13 Febraary 1952 (V.J. Tipton) (AMNH); 29, 19 Febmary 1992 (V. Roth) (CASC). Mandraka, 18°56^S, 47°56'E, 19, 10 March 1994 (W.J. Pulawski) (CASC). Ranomafana, I. Fanadiana town, 19, 16 May 1992 (Roth) (CASC). Pisaura mirabilis (Clerck 1757): GER- MANY: Bayern, Spessart, Neuhiitten, Zilch leg., 19, 19 June 1949 (AMNH ex SMFD). Ransonia mahasoana Blandin 1979: MADAGASCAR: Fian- arantsoa Province, PN. Ranomafana, Vohipara, 2ri4'S, 47°24'E, 900 m, 29, 5-7 December 1993 (Scharff, Larcher, Griswold, Andriamasimanana) (currently CASC). Antananarivo Province, Amboh- imanga, 18°44^S, 47°34'E, 1400 m, 2(3 1 1 9 , 1 No- vember 1993 (Coddington, Larcher, Griswold, An- driamasimanana, Scharff) (currently CASC). Antsiranana Province, Marojejy Reserve, 8.4 km NNW Manantenina, 14°25'S, 49°45'E, 700 m, sev- eral 6 9, 10-16 November 1993 (currently CASC). Rothus purpurissatus Simon 1898: KENYA: Lake Nakuru, N.P. campsite in Yellow Fever Forest, 1 9 , 14 May 1975 (Penniman) (AMNH). SOUTH AF- RICA: Natal, Lake Midmar, 1 (33 9 , 6 January 1991 (V.D. & B. Roth) (CASC). Tallonia picta Simon 1889: MADAGASCAR: Province Antsiranana: Nosy Be, Lokobe Forest, 13°24'58.8"S, 48°18'26.5"E, 49, 11-14 August 1992 (V.D. & B. Roth) (CASC). Montague d’Ambre, 12°30'57"S, 49°ir04"E, 29, 12 August 1992 (V.D. & B. Roth) (CASC). Thalassiopsis vachoni Roewer 1955: MADAGASCAR: Maroantsetra, 16, SMFD RII/ 10552/102 (type-label carries an invalid manuscript name). Vuattouxia Blandin 1979 sp.: MADAGAS- CAR: Toamasina Province, P.N. Perinet, near An- dasibe, 1000 m, 18°56'S, 48°24'E, 4c35 9, 4-5 No- vember 1993 (Coddington, Larcher, Griswold, Andriamasimanana, Scharff) (currently CASC); 18°55'S, 48°25'E, 1(319, 1-3 August 1992 (V.D. & B. Roth) (CASC). Chutes de la Mort, 1(3, 10 November 1959 (E.S. Ross) (CASC). Fianarantsoa Province, P.N. Ranomafana, Talatakely, 21°15'S, 47°25'E, 900 m, 59, 5-7 December 1993 (Scharff, Larcher, Griswold, Andriamasimanana) (currently CASC). Phylogenetics. — The pisaurine genera Charminus, Cispius, Tetragonophthalma, Af- ropisaura, Perenethis, Maypacius, and Poly- boea form the monophyletic Perenethis genus group, within the Pisaurinae as here defined. The synapomorphy for this genus group is the membranous saclike copulatory duct forming two large loops at least in some species of each of the seven genera except in Cispius. The genera Charminus and Cispius, them- selves sister taxa, were designated as outgroup during the cladistic analysis (see above under METHODS). The sistergroup relationship be- tween Cispius and Charminus is supported by the procurved carina, which is unique within the Pisaurinae. The outgroup genera were not newly revised for the present study and the character states of the type species of each genus were used in the cladistic analysis. However, most character states (Table 4) are identical in all known species of each genus as illustrated in Blandin’s (1978a) revision of the genera. Characters of the internal female copulatory organ {cha 13-17) and a few char- acter states in the male organs {ch 31, 33, 34), 370 THE JOURNAL OF ARACHNOLOGY not illustrated by Blandin, were assessed in two species in Charminus (C camerunensis and C. aethiopicus). In Cispius, the male char- acters were assessed in three different species (C thorelli, C. problematicus and C. biden- tatus), the female characters are based on the type species alone (C variegatus). Character description: Somatic characters. Character 0: Number of chehceral teeth; 0 = three teeth, 1 = four teeth (with occasional oc- currence of three teeth at one of the chehcerae), 2 = two teeth. Three cheliceral teeth is the most common and widely distributed character state in the Pisaurinae and thus assumed to be the primitive condition (see Table 2). However, the character is somewhat homoplastic within the Pisaurinae, e.g., the state two teeth occurs in Cispius bidentatus (Lessert 1936), all other species of the genera Charminus and Cispius have three cheliceral teeth. Character 1 : Size of cheliceral teeth; 0 = unequal, 1 = equal. The character displays homoplasy within this data set and outside (e.g., Charminus ambiguus (Lessert 1925) has three equally-sized teeth). Character 2: Shape of AER; 0 = recurved, 1 = straight, 2 = procurved, 3 = strongly pro- curved [= St proc I in Table 2], 4 = extremely procurved [= st proc 11 in Table 2]. The mor- phological difference in the AER in Maypacius and Tetragonophthalma justifies a separate coding (see Blandin’s figures, 1974a, figs. 1, 4). Character 3: Size ratio of ALE to PME; 0 = ALEPME. Character 4: Size ratio of ALE to AME; 0 = ALEAME. Character 5: Patella spina- tion; 0 = patella with one dorsal apical spine as in Charminus (Table 6), 1 = patella with two dorsal and two lateral spines as in Afro- pisaura, 2 = patella of legs I and II without spines, patella II and IV with a single dorsal apical spine, as in Tetragonophthalma. Char- acter 6: Spine length, expressed as ratio of length of tibial spine of second ventral spine- pair to width of tibia; 0 = 3, 1 = 1-1.5, 2 == 6.5. Extremely long (state 2) or short spines (state 1) are not as common in Pisauridae as moderate spine length (state 0). Female copulatory organ: Character 7: Tra- jectory of epigynal folds; 0 = anterior section of epigynal folds convergent and close togeth- er (Figs. 4, 6), 1 = anterior section of epigynal folds divergent and far apart (e.g.. Figs. 17, 20). Character 8: Carina; 0 = continuous (Figs. 4, 6, 17, 20, 23, 85, 88), 1 = separated into two distinct lateral, nearly straight branches (e.g.. Fig. 31). Character 9: Carina form; 0 = recurved (Figs. 17, 20, 23, 82), 1 = straight (e.g.. Figs. 31, 88), 2 = procurved as in the outgroup (Figs, 4, 6), 3 = undulated as in Tetragonophthalma (Fig. 23), 4 = un- dulated as in Perenethis symmetrica (Fig. 34). A recurved carina (state 0) occurs in many pisaurine spiders. Character 10: Posterior edge of carina; 0 = ridgelike (e.g.. Fig. 31), 1 = liplike, overhanging anterior section of epi- gynal folds (e.g.. Figs. 17, 23, 88). Character 11: Anterior edge of carina; 0 = weakly de- veloped (Figs. 4, 6), 1 = distinctly developed, ridgelike (e.g.. Figs. 17, 20, 31, 82, 85, 88). Character 12: Position of fossae in relation to copulatory opening; 0 = directly above cop- ulatory openings (Figs. 17, 20, 23), 1 = above and lateral to copulatory opening (e.g.. Figs. 4, 6, 42, 45), 2 = above and between copu- latory openings (Figs. 82, 85, 88). Fossae di- rectly above the copulatory openings appear to prevail in the Pisaurinae, thus this state is coded as 0. Character 13: Copulatory duct; 0 = membranous (e.g.. Figs. 532, 35, 89), 1 = posterior section sclerotized (Figs. 83, 84), 2 = anterior and posterior sections sclerotized with membranous middle section (Figs. 18, 21), 3 = fully sclerotized over its entire length {Tetragonophthalma, Fig. 23). Character 14: Number of loops of copulatory duct; 0 = sin- gle loop (Figs. 7, 83, 86), 1 = two loops (e.g., Figs. 5, 32, 35, 89), 2 = two and one half loops (Figs. 18, 21, 24). Character 15: Copu- latory duct loop sizes; 0 = first loop larger than second loop (e.g.. Figs. 5, 49, 89), 1 = first loop equal to second loop (Figs. 24, 35). Character 16: Position of head of spermathe- ca; 0 = pointing laterally (e.g., Figs. 5, 19), 1 = pointing anteriorly (e.g.. Figs. 7, 22), 2 = bent (Fig. 33). In the bent position, the hs points anteriorly, but the stalk immediately behind it is bent, thus the spermathecal duct does not run straight in that section. Character 17: Spermathecal base; 0 = with large lumen (Figs. 7, 22, 38, 41, 44, 47, 50), 1 = with small lumen, does not fill base of spermatheca (Figs. 5, 33, 35, 90), 2 = base filled with loops of spermathecal duct (Figs. 19, 24). Displays homoplasy within the genera Afropisaura, Perenethis, and Maypacius. Male copulatory organ: Characters for the male of Maypacius kaestneri were taken from SffiRWALD— PHYLOGENY OF PISAURINE SPIDERS 371 Table 4. — Character scoring. The character matrix does not contain autapomorphies of terminal taxa (e.g., flap at the conductor in Charminus), unless they are part of a multistate series. Non-applicable character states indicated by Unknown character states indicated by Somatic characters 0) Cheliceral teeth #: 3; 4(3); 2 1) Chel. teeth size: unequal; equal 2) AER rec; str; proc; strongly proc; extremely proc 3) ALEPME 4) ALEAME 5) Patellar spines: 1; 4; 0 and 1 6) Spine length to tibia width: 3; 1-1. 5; 6.5 Female characters 7) anterior epf: convergent; divergent 8) Carina: continuous; two branches 9) Carina: rec; str; proc; Tetragonophthalma; P. symmetrica 10) Posterior edge of ca: ridge; lip 11) anterior edge of ca: weak; strong 12) position of fossae: directly above copulatory openings; lateral; between 13) copulatory duct: membraneous; posterior section sclerotized; posterior and anterior section sclerotized; fully sclerotized 14) cd loops: 1; 2; 21/2 15) cd loop sizes: 1 > 2; 1 = 2; 16) hs position: lateral; anterior; bent 17) bs: large lumen; small lumen; lumen filled with duct loops Male characters 18) rta shape: round-perpendicular; flat-forward 19) rta tip: pointed; rounded 20) tibial venter: smooth; with hump 21) tegulum base: smooth; with peak 22) c tip: long; short 23) c midpiece: long; short 24) c tip: broad round; slender round; spiral; modified spiral 25) c base: narrow; broad 26) c mesally: smooth; with hump 27) c tip: with ridge-like fold; smooth 28) c tip: additional guiding lamella absent/present 29) db of da smooth; with pit 30) scl A size: small; median; large 31) scl A shape: triangular; rod; small, oval; forked 32) dst shape: C/i; Cf; Afro\ Tetra\ Pe\ PoMay 33) e length: long; moderate; short 34) pp length: short; Vi to %; total e length U ^ ^ S , S s ^ ^ ^ a, a, a., a. § ^ d a. ^ ^ 00001222222222 00111000001111 00123222222444 00001000002111 10122000002000 00112000000000 00001000002111 OOlllllllll?!! OOOOOlllllOPOO 22003141111210 0111000001?000 OOlllllllll?!! 11000111112222 0 3 2 2 3 0 0 1 0 2 2 2 1 1 0-00101 0 10 112 2 1 0 2 0 2 1 1 0 0 0 0 2 0 1 1 1 1 1 ? 0 0 0 0 0 0 ? 0 2 2 2 2 2 1 0 0 0 1 2 0 1 0 0 0 0 0 1 1 0 0 0 0 1 1 1 0 0 0 0 0 1 1 0 0 0 0 0 1 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 oil 0 0 1110 0 0 0 0 0 0 1 1 0 0 110 11 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11110 0 0 0 0 3 1 1 2 2 0 1 2 2 3 4 4 0 1 0 0 0 0 0 0 2 11112 2 1 1 0 1 0 2 2 0 0 0 0 2 2 2 1 1 0 0 0 2 2 1 1 1 1 1 2 2 0 0 0 0 0 2 2 0 0 0 1 1 2 2 1 1 2 3 3 2 2 0 0 0 0 0 2 2 1 1 0 0 0 2 2 1 1 0 0 0 2 2 0 0 1 1 1 2 2 0 0 1 1 2 2 2 0 0 2 2 2 2 2 2 2 3 3 2 2 2 4 4 5 5 2 2 2 0 0 1 2 2 2 2 1 1 0 1 2 2 372 THE JOURNAL OF ARACHNOLOGY Blandin’s figure (1975a: 389, figs. 21, 22). Character 18: Retrolateral tibial apophysis (RTA); 0 = round and perpendicular to palpal tibial (e.g.. Figs. 9, 12-15, 25, 27, 97), 1 = flat and parallel to cymbium (e.g.. Figs. 54 — 59, 91). Character 19: Tip of RTA; 0 = point- ed (e.g.. Figs. 14, 59), 1 = rounded (Figs. 28, 56, 58). Character 20: Tibial venter; 0 — smooth (e.g.. Figs. 11, 15, 92, 97), 1 = with hump-shaped apophysis (Figs. 54, 55, 58, 79). Character 21: Tegulum base; 0 = smoothly rounded (e.g.. Figs. 8, 10, 25, 29), 1 = with basal protuberance at retrolateral side of unex- panded bulb (unique within the Pisaurinae, Figs. 54, 92, 97). Character 22: Conductor tip length; 0 = long (e.g.. Figs. 10, 58), 1 = short (Fig. 25). Character 23: Conductor midpiece between tip and base; 0 = long (e.g.. Figs. 13, 55, 57), 1 = short (Figs. 91, 95). Character 24: Conductor tip shape; 0 = broad and well rounded (Figs. 10, 12, 27) as in the outgroup, 1 = slender and rounded as in Perenethis (Figs. 54-58), 2 = modified, with a spiral tip as in Polyboea (Figs. 97, 98), 3 = modified as in Maypacius (Fig. 93). Character 25: Base of conductor; 0 = narrow (Figs. 55, 95, 98), 1 = broad (Figs. 26, 28). Character 26: Con- ductor mesal margin; 0 = smooth, 1 = with mesal inflatable hump as it occurs in Pere- nethis (Fig. 55). Character 27: Conductor tip folds; 0 = conductor tip with a ridgelike fold formed by the membranous inner conductor wall as in Charminus and Tetragonophthalma (Fig. 10, 12, 27), 1 = inside of tip smooth (Fig. 56). Character 28: Posterior guiding la- mella at tip of conductor; 0 = absent (Fig. 10), 1 = present (Figs. 95, 98). Character 29: Dor- sal branch of distal tegular apophysis; 0 smooth, 1 = with pit (Figs. 95, 98). The mor- phological similarity in the distal tegular apophysis between Polyboea and Maypacius is striking and links both genera. Character 30: Size of sclerite A; 0 ^ small (Fig. 29), 1 = medium (Figs. 11, 14, 15), 2 ^ large (Figs. 94, 99). This sclerite can only be seen in the expanded bulb. As far as known, a small sclerite A is widely distributed within the Pi- saurinae. Character 31: Sclerite A shape; 0 = triangular, as in Charminus and Cispius (Figs. 11, 14), 1 = rod-shaped, as in Tetragonoph- thalma (Fig. 29), 2 = small, elongated oval, as in Perenethis, 3 = large, one end forked as in Afropisaura valida, Polyboea and Maypa- cius (Figs. 91, 93, 99). Character 32: Shape of sclerotized tube of embolic division; 0 = as in Charminus, 1 = as in Cispius, 2 = as in Afropisaura, 3 — as in Tetragonophthalma, 4 = as in Perenethis, 5 = as in Polyboea and Maypacius. Character 33: Embolus length; 0 - long (Figs. 11, 25, 26, 28, 54, 57), 1 - moderate (Fig. 14, 99), 2 = short (Fig. 95). Character 34: Length of pars pendula along- side the embolus; 0 = short (Figs. 11, 99), 1 = along the embolus (Figs. 26, 28, 56), 2 == total embolus length to its tip (Figs. 14, 15, 58). Analysis: Hennig86 runs containing all characters (multistate characters unordered) resulted in three equally parsimonious, highly resolved trees of 87 steps, with a consistency index {ci) of 0.72 and a retention index (n) of 0.76, differing only in the resolution of the three species of the genus Maypacius. The equally long Nelson Consensus Tree (Fig. 2), is part of the original series of three trees, and places these three species in an unresolved tri- chotomy. An inspection of the other two, mu- tually exclusive trees presenting the three spe- cies fully resolved, showed that the resolution is based on presumed character states of miss- ing characters. Only in two of the nine species of the genus Maypacius are the males and fe- males known, and few specimens are avail- able for study (see below under Maypacius), causing this lack of data. Optimization of character- state changes may differ with the choice of optimization schemes, ACCTRAN (depicted here in Fig. 2) or DELTRAN. In the following description, only non-homoplastic character- state changes supporting a clade un- der both optimization schemes are discussed, unless noted otherwise. The Ingroup (clade A), containing the gen- era Tetragonophthalma, Afropisaura, Pere- nethis, Maypacius, and Polyboea is defined by the following synapomorphies of non-homo- plastic character- state changes: Procurved an- terior eye row {ch 2), the shape of epigynal folds {ch 7), the strongly developed anterior edge of the carina {ch 11), and a pars pendula Vi-Ya along the embolus {ch 34). The genera Afropisaura and Tetragonophthalma (clade B) form the sister-group to clade C. The sister- group relationship between Afropisaura and Tetragonophthalma is supported by five non- homoplastic character- state changes: The po- sition of the fossae above the copulatory openings {ch 12), the undulated posterior sec- SffiRWALD— PHYLOGENY OF PISAURINE SPIDERS 373 Table 5. — Character performance. * Different op- timizations ACCTRAN/DELTRAN. Character number Character states Steps ci ri 0) m(3) 2 100 100 1) b 2 50 83* 2) m(5) 4 100 100 3) m (3) 3 66 66* 4) m(3) 4 50 33 5) m(3) 2 100 100* 6) m(3) 3 66 66* 7) b 1 100 100 8) b 1 100 100 9) m (5) 5 80 66* 10) b 2 50 66 11) b 1 100 100 12) m(3) 2 100 100 13) m (4) 4 75 50* 14) m(3) 3 66 75 15) b 2 50 0 16) m(3) 5 40 40 17) m (3) 5 40 40 18) b 2 50 75 19) b 2 50 50 20) b 1 100 100 21) b 1 100 100 22) b 1 100 100 23) b 1 100 100 24) m(4) 3 100 100* 25) b 1 100 100 26) b 1 100 100 27) b 2 50 80 28) b 1 100 100 29) b 1 100 100 30) m(3) 3 66 75* 31) m (4) 4 75 75* 32) m (6) 5 100 100* 33) m(3) 3 66 50* 34) m (3) 4 50 0* tion of the copulatory duct {ch 14), copulatory duct with IVi loops), spermathecal base filled with loops of spermathecal duct (ch 17, re- versal in Afropisaura ducts), the broad base of the conductor (ch 25), and the rod-shaped sclerite A (ch 31). Non-homoplastic apomor- phies for the genus Afropisaura are the spi- nation of the patella (ch 5), the partly sclero- tized copulatory duct (ch 13), the short pointed conductor (ch 22), and — under DEL- TRAN optimization— the peculiar shape of the distal sclerotized tube of the embolic di- vision (ch 32). Clade C (sister taxon to clade B), containing the genera Perenethis, Poly- boea and Maypacius is supported by the loss of one cheliceral tooth (ch 0), the straight Ca- rina (ch 9, with special form in P. symmetrica and recurved carina in M. petrunkevitchi Les- sert 1933), the bent spermathecal head (ch 16, with reversals in P. symmetrica and M. pe- trunkevitchi), and the basal protuberance at the tegulum (ch 21). The sister-group relation- ship of Polyboea and Maypacius (clade D) is corroborated by mesal position of the fossae (ch 12), the additional guiding lamella in the conductor (ch 28), the pit in the dorsal branch of the distal tegular apophysis (ch 29), the large, forked sclerite A (ch 30), and — under ACCTRAN optimization-— the shape of the distal sclerotized tube of the embolic division (ch 32). The species of Maypacius included in this study are defined by the strongly procur- ved anterior eye row (ch 2), the short conduc- tor midpiece (ch 23), and — under ACCT- RAN— the short embolus (ch 34). The genus Perenethis, the sister taxon of clade D, is defined by four apomorphies of non-homoplastic character-state changes: Ca- rina with two branches (ch 8), the ventral tib- ial apophysis (ch 20), the slender, rounded tip of the conductor (ch 24), the mesal hump at the conductor (ch 26), and — under BEL- TRAN— the shape of the distal sclerotized tube of the embolic division (ch 32). The sis- ter-group relationship of the African species, Perenethis simoni and Perenethis symmetrica, is weakly supported by a homoplastic char- acter-state change, the rounded tip of the re- trolateral tibial apophysis (ch 19). The Asian- Australian species of the genus are weakly supported by the large lumen in the sperma- thecal base (ch 17), a very homoplastic char- acter throughout the Pisaurinae. The dado- gram clearly demonstrates that the astounding phenetic similarity between P. simoni (Lessert 1916), P. sindica (Simon 1897) and P. ven- usta L. Koch 1878 is based on symplesiom- orphies alone and suggests, however weakly, that the African species and the Asian species of Perenethis form separate clades. Zoogeography. — ^Pisaurine genera are pre- dominantly African (south of the Sahara), but one genus, Polyboea, is restricted to south- east Asia. Pisaura itself occurs in Europe and Asia. The distribution patterns within the Pi- saurinae display an interesting peculiarity: Three genera, Perenethis, Dendrolycosa (ge- nus unrevised, Blandin 1979a, figs. 30, 34, 36) 374 THE JOURNAL OF ARACHNOLOGY Table 6. — Leg spination patterns of Charminus, typical for the Pisaurinae. Genus specific variations shown below. Abbreviations: 1 = average size spine; i = short spine; I = long spine; * = spine dislocated to retrolateral; v = spine dislocated to ventral; 2 = spine pair, average length; ii = spine pair, spines short. Notation indicates location of spine on leg segment, e.g., proximal, apical, and to other spines of segment. Spine length [ventral tibial spine, second pair, first leg]: spine length : tibia width == 3 set as normal length. Charminus leg do Fe 1 1 1 Pa 1 Ti 1* 1 I pi i 0 0 1 1 1 rl 1 111 1 1 1 V 2 2 2 ii leg d Fe 1 1 1 Pa 1 Ti 1* 1 II pi 1 111 1 1 1^ rl 1 111 1 1 1 V 2 2 2 ii leg do Fe 1 1 1 Pa I Ti 1* 1 III pi 1 111 1 1 1^ rl 1 111 1 1 1 V 2 2 ii leg do Fe 1 1 1 Pa 1 Ti 1* 1 IV pi i ill 1 1 1^ rl 0 0 0 1 1 1 1 V 2 2 ii Afropisaura leg do Fe 1 1 1 I-IV Pa i 1 1 pi 0 111 1 i rl 1 111 1 i Tetragonoph thalma leg do Fe 1 1 i I, II Pa : 0 III, IV Pa I pi i 111 1 rl 1 111 1 9 leg do Ti 0 III, IV Ti 1* 1 I, II pi 1 iv 1 1 rl 1 1 1 1 V 2 2 2 ii 2 2 ii 6 leg do Ti d") 1 III, IV Ti 1* 1 pi 1 1^ 1 1^ rl 1 1 1 1 V 2 2 2 ii 2 2 ii Polyboea leg do Fe 1 I i I»IV Ti 1* 1 I pi 0 d 1 1^ rl 1 111 1 1 1 Me Me Me Me 11 111 111 2 2 i ii 111 111 2 2 i ii 111 111 2 2 i ii 111 111 2 2 i 2 2 2 0 V SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 375 Figure 2. — Preferred Cladogram (identical to Nelson Consensus Tree). ACCTRAN character optimi- zation; character mapping: black rectangles = non-homoplastic character state origination, white rectangles = homoplastic character state origination [implemented in CLADOS by setting homoplasy indicator, HOM=0]. and Euprosthenops (Blandin 1975b), contain both African and Asian species. In the genus Perenethis, two species, P. simoni and P. symmetrica, occur in Africa (Fig. 3); three species, P. sindica, P. denti- fasciata and P. fascigera are distributed in Asia (including India and Sri Lanka); and one species, P. venusta occurs in Australia and the Bismarck-Archipelago. Maypacius, the sister taxon of the Asian genus Polyboea, is restricted to Africa. This specific pattern, the distribution of closely related species in Africa south of the Sahara and Asia/ Austra- lia, can be found in other Pisauridae sensu lato as well. The pisaurid genus Thalassius, presumably a relative of the worldwide genus Dolomedes, shows a similar pattern. Eight species of the genus Thalassius occur in Af- rica, and four species in Madagascar. A sin- gle species, T. jayacari EO. Pickard-Cam- bridge 1898, is found in the Middle East; and two species, T. albocinctus (Dolescall 1859) and T. phipsoni EO. Pickard-Cambridge 1898, are distributed in south-east Asia, in- cluding China, the Philippines and Singapore (Sierwald 1987). There are other groups dis- playing an African-Southeast Asian distri- bution of close relatives, e.g., the tropical 376 THE JOURNAL OF ARACHNOLOGY tree-frog subfamily Rhacophorinae (family Ranidae) (Savage 1973), the genus Francol- inus among the galliform birds (Dinesen et al. 1994), several members of the bird family Rallidae (Olson 1973), and members of the plant family Sapotaceae (Pennington 1991). Africa, South America and India (together with Madagascar) formed a contiguous land mass about 155 million years before present. Madagascar separated from Africa over 120 m.y.b.p. and South America separated from Africa about 95 m.y.b.p. (Smith et al. 1994). Since no member of the Perenethis genus group has been recorded from Madagascar or the Americas so far, it is unlikely that the or- igin of the terminal taxa studied here dates back more than 100 million years. When op- timizing the current distribution of the termi- nal taxa like a character on the proposed cladogram (Fig. 3) it is most parsimonious to assume an African origin for the group. Using Bremer's (1992) method to reconstruct the an- cestral area of the group under study here, yields the same result Assuming an African distribution of early pisaurine taxa requires a hypothesis regarding the presence of Polyboea and Perenethis in Asia and Australia. Two in- dependent events, one in clade D and another within Perenethis, have to be proposed to ac- count for the current distribution pattern. Ac- cording to Axelrod & Raven (1978), lowland rain forest and subtropic rain forest covered most of Africa during the Paleocene (60 m.y.b.p.), thus ancestors of the group under study were likely to be more widely distrib- uted in Africa than today, where they are mostly restricted to Africa south of the Sahara. At that time, the Tethys Sea was an effective barrier between Africa and Asia. In the mid- Miocene (15 m.y.b.p.) a land bridge formed between Africa and Arabia, connecting the African land mass with Asia. During the Mio- cene, the exchange of mammalian taxa be- tween Africa and Asia increased dramatically presumably via this land bridge, resulting in a significant decrease of mammalian taxa en- demic to Africa in the Pliocene. A two-way traffic via the Arabian Peninsula affected the composition of both faunas in Asia and Africa (Magglio 1978). The habitat conditions along that passage-way were suitable at various times during the Cenozoic for many large mammals, indicating the existence of vegeta- tion cover. Savage (1973) explains the Afri- can~Asian distribution of the tropical tree-frog subfamily Rhacophorinae (Family Ranidae) as an immigration from Africa to Asia during the early Cenozoic. Dinesen et al. (1994) also as- sume that the African-Asian distribution of several related forest birds can be explained by alternations of isolation and range expan- sion opportunities via an African- Arabian land bridge to Asia. The blockage of the Tethys Sea by a land bridge between Africa (including Arabia) and Asia in the mid-Miocene (15 m.y.b.p.) with subsequent alteration of the latitudinal air- ocean circulation produced a drier climate in northern Africa (Crowell & Frakes 1970) and caused the expansion of savanna and sclero- phyll vegetation over northern Africa and the Sahara region (Axelrod & Raven 1978). The hot, dry Saharan-Libyan desert did not devel- op before the Pleistocene. To my knowledge, no Pisauridae sensu lato have been reported from very arid habitats. Judging from their current habitat preferences, deserts and high mountainous areas may be the most difficult areas to cross, thus presenting effective bar- riers. The restriction of most pisaurine taxa to the south of the Sahara supports this notion. At least two independent events have to be assumed for the Perenethis genus group to ac- count for the occurrence of species of this group in Asia. Instead of invoking hypotheses of chance dispersal (e.g., ballooning) over pre- existing barriers like the Sahara, range expan- sions of the respective ancestral species by tracking the expansion of suitable habitats across the mid-Miocene (15 m.y.b.p.) land bridge between Africa (including Arabia) and Asia via Iran before the development of the Sahara may be more plausible. From the end of the Pliocene until today, the expansion of the Sahara formed an increasingly effective barrier to animal migration (Coryndon & Sav- age 1973). Thus, two independent range ex- tension events are proposed, most likely be- tween the mid Miocene and beginning of the Pleistocene, during the emergence of clade D and the emergence of the genus Perenethis which caused the African-Asian distribution of these clades. TAXONOMY Discriminatioe of morpho-species.— ^ When working on the alpha-taxonomic level of poorly known groups for which only pre- SffiRWALD— PHYLOGENY OF PISAURINE SPIDERS 377 Africa Africa Africa Africa Africa South-East Asia Africa Africa Africa Africa Africa India/Pakistan India Australia Figure 3. — Area cladogram, taxon names replaced by their distribution. served material is available, the discrimination of “morpho-species” as a reasonable assess- ment of biological species becomes the prac- tical solution. In the species revisions of the genera Perenethis and Tetragonophthalma be- low, I diverge in some cases quite remarkably from species delimitations drawn by previous authors, mainly Roewer (1955), but also Blan- din (1974a---1979b). For this reason it appears to be appropriate to present the rationale for morpho-species delimitations employed in this study. A species-typical trait (== character or character state) should: a) occur consis- tently in all members of the species, b) be qualitatively discrete {= variation should not show continuous overlap among organisms assigned to different species), and c) prefera- bly show congruence with at least one other such trait. Studies on individual variability within samples and/or populations, whenever feasible, should be considered. In most spider groups, discrete traits in the male palpal or- gan, e.g., shape of median apophysis or tibial apophysis, are good indicators for the recog- nition of distinct species. Which structural part of the male palpal organ reveals species- typical attributes varies among closely related spider groups, e.g., at the generic level. The retrolateral tibial apophysis, the median apophysis, the conductor and, within the Pi- saurinae, the distal tegular apophysis, are of- ten species-typical. The female epigynum can show quite a range of variation within a species, which led 378 THE JOURNAL OF ARACHNOLOGY to numerous redundant species descriptions, e.g., Thalassius spinosissimus (Karsch 1878) with its 40 junior synonyms (Sierwald 1987). The ducts and spermathecae of the vulva show less individual variability and are therefore more reliable features for the discrimination of species, at least in all Pisauridae I have ex- amined so far, e.g., the shape of the head of spermathecae in Thalassius (see Sierwald 1987). For alpha-taxonomic studies of spiders, the structure of the vulva should always be included if females are available. Within the species studied for the present paper, the duct leading from the head to the base of the sper- mathecae may show variations in its loops, even differing between the left and right sides of a single individual. Subadult females with heavily sclerotized pre-epigyna (primordia of the developing copulatory organs, Sierwald 1989b) have been mistaken for adults and consequently caused the description of syn- onymous taxa or nomina dubia as in Maypa- cius bilineatus (Pavesi 1895), see Blandin 1975a. Congruence of discrete traits in both male and female copulatory organs permit re- liable species-discrimination. Somatic fea- tures, such as color-pattern, eye-pattem, spi- nation and leg formula that occur consistently in concordance with discrete traits of copula- tory organs of one of both sexes furnish ad- ditional species-typical attributes. Somatic features often facilitate recognition of conspe- cific sexes, but may display sexual dimor- phism (see Tetragonophthalma below). From my experience with Pisauridae, coloration and color-pattern are prime candidates for sexual dimorphism and polymorphism within spe- cies. However, there are noteworthy excep- tions such as the color pattern in both sexes of Perenethis symmetrica which is consistent- ly different from all other species of the ge- nus. Taxonomic history of the Perenethis ge- nus group. — The Perenethis genus group comprises the pisaurine genera Charminus, Cispius, Tetragonophthalma, Afropisaura, Perenethis, Maypacius, and Polyboea. The African genera Charminus and Cispius are morphologically similar and have been con- fused in the past. Blandin (1978a) separated both genera conclusively by features of the eye pattern, the tibial apophysis and the em- bolus, and moved several species originally described in Cispius to Charminus. The gen- era Tetragonophthalma, Perenethis and May- pacius each have caused taxonomic problems and misinterpretations since their introduction to arachnology (comprehensive review by Blandin 1974a). Several species were shifted mainly among the three genera [e.g., Maypa- cius bilineatus, described sub Tetragonoph- thalma], probably due in part to the homo- geneity of certain morphological features (especially the structure of the epigynum) and in part due to misidentification of genus-typ- ical characters (e.g., retromarginal cheliceral teeth). The major event causing confusion was an incorrect re-description of the genus Tetra- gonophthalma by Simon (1898a), in which he cited two retromarginal cheliceral teeth. Karsch (1878), in the original description of Tetragonophthalma, did not give the number of teeth at the posterior margin of the chelic- erae. Dahl (1908) examined the holotype of Tetragonophthalma phylla Karsch 1878 (type species of the genus), noted four cheliceral teeth, and rejected Simon’s synonymy of Per- enethis with Tetragonophthalma. Meanwhile, Simon (1898) had introduced the genus Pha- laea to accommodate species with four teeth at the posterior margin of the chelicerae, thus creating a junior synonym for Tetragonoph- thalma. Subsequent authors either followed Simon’s interpretation of the genus Tetragon- ophthalma [e.g., Bosenberg & Strand 1906 in the description of Perenethis fascigera under Tetragonophthalma] or rejected it (Pocock 1900). Roewer (1955) reviewed the issue and re-examined the immature type of T. phylla (which apparently has been lost since; fide Blandin 1976a). The genus Afropisaura was recently described by Blandin (1976b) for three African species formerly placed in the genus Pisaura. The monotypic genus Poly- boea was introduced by Thorell (1895) for a juvenile male from Burma. Male and female of the type species are figured here for the first time. Charminus Thorell 1899 Figs. 4, 5, 8-11 Charminus Thorell 1899: 83. Type species, by orig- inal designation, Charminus camerunensis Tho- rell 1899: 83; c??, Cameroon. Nine species are currently recognized in the genus; two species are known from females only, one is known from a single male. Valid species of the genus: Charminus aethiopicus SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 379 Caporiacco 1939, S' $ known; Cispius ambig- uus Lessert 1925, S $ known; Cispius ato- marius Lawrence 1942, S ? known; Charmi- nus bifidus Blandin 1978 (1978a), S known; Charminus camerunensis Thorell 1899, S 9 known; Cispiolus marfieldi Roewer 1955, S 9 known; Cispius minor Lessert 1928, S 9 known; Cispius natalensis Lawrence 1947, 9 known; Charminus rotundus Blandin 1978 (1978a), 9 known. Diagnosis.— "AER recurved {ch 2), three cheliceral teeth {ch 0); epigynal folds anteri- orly parallel and close to each other {ch 7), Carina procurved {ch 9); conductor with nar- row basal stalk {ch 25) and broad, rounded, flaplike apical section {ch 24); single guiding fold for embolus {ch 27); embolus long and whip like {ch 33). Autapomorphic characters: AE nearly same size as PE {ch 3), lateral flap at conductor, carina with lateral lobes. Syna- pomorphic characters: Procurved carina as in Cispius {ch 9). The membranous saclike cop- ulatory duct forming two loops {ch 14) occurs at least in some species of the following pi- saurine genera: Afropisaura, Perenethis, and Polyboea; in Tetragonophthalma, the two loops are sclerotized {ch 13) but their shape and trajectory are identical to Afropisaura. The copulatory ducts in other pisaurine genera are different, as far as currently known. Description of Charminus camerunensis Thorell 1899.— (2^“ 1 $). Measurements: 9 slightly larger than S , S with longer legs than 9. 9 body 7. 8-9. 7 long, prosoma 2.8-3. 5 long, 2.5-2.9 wide. Leg length: (prosoma 3.5 long) Fe 5.2, PaTi 6.5, MeTa 7.5, total length 19.5. S body 8.7 long, prosoma 3.5 long, 2.7 wide. Leg length: Fe 6.3, PaTi 8, MeTa 9.8; total length 24.1. Eye pattern: AER recurved and only slightly shorter than PER, AE slight- ly smaller than PE, PME:AME = 1.2; ALE slightly smaller than AME or AME^ALE. Chelicerae: Posterior margin dentition some- what variable even within individuals, mostly with three, some individuals with four teeth; teeth unequal in size, equally-sized in Char- minus ambiguus). Spine pattern: See Table 6. Epigynum (Fig. 4): Epigynal folds parallel and close together anteriorly, curved in the middle section, adjoining in the posterior sec- tion; entire carina procurved; posterior edge ridgelike, anterior edge indistinct; fossae lat- eral to copulatory openings. Vulva (Fig. 5): Copulatory duct membranous, two loops, first Figures 4, 5. — Charminus camerunensis from Gabon. 4, Epigynum; 5, Vulva. Scale line = 0.5 mm. loop larger than second, head of spermatheca pointing laterally, spermathecal duct with one loop, base of spermatheca bulbous, with small lumen. Male palp (Figs. 8-11; Sierwald 1990, figs. 49-50): Retrolateral tibial apophysis sim- ple, perpendicular, tip pointed; conductor base narrow, apical section broad with genus-typi- cal flap (Blandin 1978a: figs. 22-27) and sin- gle low guiding fold; median apophysis with S-shaped hook; distal tegular apophysis with wing and hook; base of dst with two ridges, embolus long, pars pendula of embolus length. Taxonomic note. — Charminus aethiopicus (Caporiacco 1939) NEW COMBINATION. Cispius novus Caporiacco 1941 and Cispius tertali Caporiacco 1941 are both subjective ju- nior synonyms of Charminus aethiopicus (Ca- poriacco 1939) NEW SYNONYMIES. The 380 THE JOURNAL OF ARACHNOLOGY male palps of C. tertali (AME {ch 4); three equal- ly-sized cheliceral teeth {ch 0), posterior sec- tion of copulatory duct of vulva with several, partly sclerotized undulations {ch 13). Auta- pomorphic characters: Truncus attached to distal sclerotized tube of embolic division, forming an angle; large median apophysis with blade- shaped tip; short conductor with broad base, tapering apically {ch 22). Central excavation opening posteriorly under lip-like Carina. Vulva: Anterior section of copulatory duct sclerotized {ch 13). Synapomorphic char- acters: Liplike carina {ch 10), sclerotized an- terior section of copulatory duct, and posterior section of copulatory duct undulated as in Tet- ragonophthalma {ch 13). Description of A. valida and A, duds. — Measurements: A. duds: Both sexes of same size, males with longer legs than females. Fe- male range from body 9.75 long [prosoma 3.5 long, 3.0 wide (SMFD RII/7930/52)] to body 13.49 long [prosoma 5.16 long, 4.33 wide (SMFD RII/ 10008)]. Male range from body 11.6 long [prosoma 4.58 long, 3.70 wide (ho- 382 THE JOURNAL OF ARACHNOLOGY Figures 8-11. — Charminus, left male palp. 8, 9, Charminus aethiopicus from Kenya (syntype; MZUF). 8, Unexpanded palp, ventral view; 9, retrolateral tibial apophysis. Scale line = 0.5 mm. 10, 11, Charminus camerunensis from Cameroon (syntype; NHRS, 1406). 10, Unexpanded palp, ventral view; 11, Expanded palp, prolateral view. Scale line = 0.5 mm. J Figures 12-16. — Cispius, left male palp. 12-14, Cispius thorelli from Zaire (MRAC 148.599). 12, Unexpanded palp, ventral view; 13, Unexpanded palp, retrolateral view; 14, Expanded palp, ventral view (MRAC 148.599). 15, 16, Cispius bidentatus from Zaire (MRAC 145.399). 15, Expanded palp, ventral view; 16, Embolus sclerites, schematic. Scale line = 0.5 mm. SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 383 Figures 17-19. — Afropisaura valida from Congo (MRAC 29.531). 17, Epigynum; 18, Vulva; 19, Left spermatheca, dorsal view. Scale line: 17, 18 = 0.5 mm; 19 = 0.2 mm. lotype ducisyi to body 13.1 long [prosoma 5.2 long, 4.16 wide (SMFD RII/ 103 29/79)]. A. valida: $ range from body 9.3 long [prosoma 4.16 long, 3.66 wide (MRAC 29 528-31)] to body 17 long [prosoma 7.08 long, 6.25 wide (lectotype, MNHN 4922)]. Leg length: (pro- soma 4.16 long) Fe 4.41, PaTi 5.6, MeTa 5.41; total length 15.42. Male range from body 11.26 long [prosoma 4.66 long, 3.83 wide (MRAC 29647)] to body 12.88 long [prosoma 5.8 long, 4.5 wide (MNHN “neallotype”)]. Leg length (prosoma 4.66 long) Fe 5.83, PaTi 7.5, MeTa 7.16; total length 20.5. Eye pattern: Figures 20-22. — Afropisaura duds from Zaire (SMF RII/10008). 20, Epigynum; 21, Vulva; 22, Left spermatheca, dorsal view. Scale line: 20, 21 = 0.5 mm; 22 = 0.2 mm. AER: Straight in A. valida, procurved in A. duds', PME>PLE=ALE>AME {ch 3,4), A. valida: PME:AME - 1.2, AME=ALE; A. duds: PME:AME = 1.6, AME:ALE = 0.7; in A. duds and A. rothiformis AME conspicu- ously smaller (Blandin 1976b, figs. 4, 19, 20) 384 THE JOURNAL OF ARACHNOLOGY Figures 23, 24. — Tetragonophthalma vulpina from Zaire (MRAC 12.668). 23, Epigynum; 24, Vulva. Scale line = 0.5 mm. than in A. valida. Chelicerae: Posterior margin with three equally-sized teeth, equally spaced. Spine pattern (Table 6): Patella with a thin dorsal proximal spine and two lateral spines. Spine length [ventral tibial spine, second pair, first leg]: Spine lengthitibia width — 3. Epi- gynum (Figs. 17, 20): Rather large in all three species; epigynal folds form a V; fossae above copulatory openings and in A. valida closer together than in A. duds; deep pit under me- dian section of carina opening posteriorly; copulatory openings in A. valida considerably larger than in A. duds. Vulva (Figs. 18, 19, 21, 22): Copulatory duct divided in three sec- tions; anterior section (cdl) sclerotized with large lumen; middle section (cd2) membra- nous saccate tube, forms two loops of unequal size; posterior section (cd3) narrow fold with sclerotized edge; head of spermatheca of mod- erate size, pores conspicuous; position of head of spermatheca appears to be variable within A. duds pointing anteriorly or posteriorly (type of Pisaura camerunensis [= A. dudsY. Head of spermatheca seems to point posteri- orly, visible through body wall); A. valida — specimen examined: Head of spermatheca points laterally. A. valida (Fig. 19): Sperma- thecal duct with four loops; base of sperma- theca without lumen. A. duds (Fig. 22): Sper- mathecal duct forms single curve, base of spermatheca with large lumen. Male palp (Figs. 25, 26, based on A. valida and A. duds, Blandin 1976b, fig. 26): Tegulum platelike and strongly sclerotized on both sides; con- ductor stiff, sclerotized, with broad base and pointed tip, considerably shorter than in all other Pisaurinae; prolateral wall of conductor membranous and inflatable; median apophysis flat with bladelike tip, pointing retrolaterally; distal tegular apophysis large, with club- shaped base, ventral branch ends in pointed hook. Sclerite A well developed with two prongs pointing dorsally in A. valida, simple saber-shaped in A. duds (not visible in unex- panded palps). Distal sclerotized tube of em- bolic division large; truncus of embolus arises at its distal tip and runs backwards, thus form- ing a sharp angle (called “protuberance” by Blandin). Broad pars pendula follows truncus about Va of embolus length; concave edge of pars pendula sclerotized, especially proxi- mately. Natural history. — Females of Afropisaura valida construct a nursery web (Blandin 1979b). Specimens examined. — A. duds: ZAIRE: Kivu Province, Lake Kivu, S holotype, (ZMHB 28 356). TANZANIA: Arusha, 9 “allotype” 2c? (SMFD RII/10329/79). ZAIRE: Upemba Nat. Park, 1619 (SMFD RII/10008). CAMEROON: Yaounde, 1 $ (holotype Pisaura camerunensis) (SMFD RII/7930/ 52). A. rothiformis: NIGERIA: Abarka (Kwale) Warri, 19,2 January 1949 (B. Malkin) (CASC). CAMEROON: Mkuyka, Victoria Div., Ic?, 24-29 June 1949 (B. Malkin) (CASC). ANGOLA: Lunda Province, Dundo, 19, 21 September 1949 (B. Malkin) (CASC). A. valida: CONGO: lc?29, (MRAC 29.528-29.531). Ic? (MRAC 29.647). IVORY COAST: Lamto, 6 “neallotype” (MNHN). SENEGAL, 9 lectotype (designated by Blandin) 2 juv. (MNHN 4.922). ANGOLA: Lunda Province, Dundo, Ic?, 21 September 1949 (B. Malkin) (CASC). SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 385 Figures 25, 26. — Afropisaura valida, left male palp from Congo (MRAC 29.647). 25, Unexpanded, ventral view; 26, Expanded, retrolateral view. Scale line = 1 mm. Tetragonophthalma Karsch 1878 Figs. 23, 24, 27-30 Tetragonophthalma Karsch 1878: 329. Type spe- cies, by monotypy, Tetragonophthalma phylla Karsch 1878: 329; $ juvenile; Ghana, Accra. Im- mature female type specimen, apparently lost (fide Blandin 1976a: 588). Considered a nomen dubium by Blandin (1976a: 588). Blandin (1976a) recognized eight valid spe- cies in the genus, placed T. ferox (Pocock 1899) in the synonymy of T. crassa (Thorell 1899), and listed five nomina dubia. Types of every available African species of the genus Tetragonophthalma (9(312$) were examined for the present study. All eight species are here considered to be conspecific. No concor- dant differences were found in the males ex- amined. The female epigynum displays a moderate range of variation, but the vulvae display only minor variability. The body- length variation of adult females is high. Therefore, the nominal species T. balsaci, Phalaea crassa, Phalaea ferox, T. guentheri, T. lecordieri, T. pelengea, Phalaea thomensis, and T. wittei are here regarded as subjective junior synonyms of Tetragonophthalma vul- pina (Simon 1898). Diapontia freiburgensis Keyserling 1877 (1877: 671), transferred to Tetragonophthalma by Keyserling (1891: 255), and Tetragonophthalma obscura Key- serling 1891 (1891: 256), the only South American species ever associated with one of the pisaurine genera as here defined, were transferred to Porrimosa Roewer 1960, family Lycosidae, by Capocasale (1982: 146). Tetragonophthalma vulpina (Simon 1898) Phalaea vulpina Simon 1898b: 14 (c? $). Phalaea crassa Thorell 1899: 80 ($) NEW SYN- ONYMY. Phalaea ferox Pocock 1899: 863 ($); considered a subjective junior synonym of crassa by Blandin (1976a: 592) NEW SYNONYMY Phalaea thomensis Simon 1909: 386 ($) NEW SYNONYMY T. guentheri Roewer 1955: 172 ($) NEW SYN- ONYMY. T. pelengea Roewer 1955: 179 ((3$) NEW SYN- ONYMY. T wittei Roewer 1955: 181 {6 9) NEW SYNON- YMY. T lecordieri Blandin 1976a: 601 (c? $) NEW SYN- ONYMY T balsaci Blandin 1976a: 602 {6 9) NEW SYN- ONYMY Diagnosis. — Large spiders ( $ up to 40 mm long) with the following autapomorphic char- 386 THE JOURNAL OF ARACHNOLOGY Figure 27. — Tetragonophthalma vulpina, left male palp, unexpanded, ventral view; (paratype of T. iecordieri; MRAC 123.720). Scale line = 1 mm. acters: AER strongly procurved (ch 2); four cheliceral equally- sized teeth (ch 0, 1); sexual dimorphism in spination of patella and tibia unique within the Pisaurinae (ch 5); spines short (ch 6); carina continuous, with deep me- dian notch; copulatory duct fully sclerotized (ch 13); hook at distal tegular apophysis in male bulb large. Synapomorphic characters: Liplike carina (ch 10), sclerotized anterior section of copulatory duct and undulated pos- terior copulatory duct shared with Afropisaura (ch 13). Description of T, vulpina. — Measure- ments: Females are larger than males, males have relatively longer legs than females. Fe- male body range from 16.1 long (prosoma 6.8 long, 5.2 wide [holotype of P. thomensis}) to 40 long (prosoma 16 long, 11.4 wide [holo- type of P. crassa])^ Leg length: Female (pro- soma 8.4 long, holotype of T. Iecordieri) Fe 13.5, PaTi 18.3, MeTa 22; total length 53.8. Male body range from 16.7 long (prosoma 6.2 long, 5 wide [paratype of T. Iecordieri]) to 24.5 long (prosoma 10 long, 8.4 wide [para- type of T. pelengea, MRAC 119.706]). Leg length (prosoma 8.2 long, [paratype of T. pe- lengea, MRAC 119.706]) Fe 17, PaTi 22, MeTa 24; total length 63. Eye pattern: AER strongly procurved; anterior lateral eyes on tu- bercles of variable size; PLE > PME = ALE > AME (ch 3,4); PME:AME = 1.1-1.48. AME:ALE = 0.7-0.9. Variability in eye-sizes reflects variability in body-size. Chelicerae: Posterior margin typically with four equally- sized teeth. Variation: Less than four teeth on one chelicera or small additional tooth or teeth of different sizes, e.g., P. thomensis: Right chelicera 3, left chelicera 4. Spine pattern (see Table 6): Patella spination on legs I and II differs from patella spination on legs III and Figures 28-30. — Tetragonophthalma vulpina, expanded left male palp. 28, 29, Ivory Coast (paratype of T. Iecordieri, MRAC 123.720); 30, Congo (paratype of T. balsaci, MNHN, Simon coll 8536). 28, Expanded, retrolateral view; note expanded conductor; 29, Expanded, ventral view; 30, Hook at distal tegular apophysis, prolateral view. Scale line: 28, 29 = 1 mm; 30 = 0.2 mm. SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 387 IV, Sexual dimorphism in dorsal spination of tibia I and 11. Spine length: Short [ventral tib- ial spine, second pair, first leg]: Spine length: tibia width = 1.5. Epigynum (Fig. 23): Epig“ ynal folds anteriorly divergent, middle section straight and far apart, posterior section adjoin- ing; Carina continuous, unique shape, forming two flaps over copulatory openings; fossae ap- proximately above copulatory openings. Vul- va (Fig. 24): Copulatory duct entirely sclero- tized, with two wide loops and a narrow posterior section; head of spermatheca point- ing anteriorly, spermathecal duct with five loops, base of spermatheca without lumen. Male palp (Figs. 27-30): Retrolateral tibial apophysis simple, perpendicular to palpal tib- ia; conductor large, tip broad with low guiding fold; median apophysis slender with hook; distal tegular apophysis with hook, without wing; sclerite A small, rod- shaped; embolus moderately long, pars pendula over Vs of em- bolus. Natural history. — Pocock (1899) men- tioned a collector’s description of a ''Tetra- gonophthalma phylla"' web. That description fits the figure given by Blandin (Blandin & Celerier 1981) for webs of the genus Eupros- thenops. Blandin (1976a) collected Tetragon- op hthalma in wooded habitats. According to my own observations (1980, South Africa, Natal, Hluhluwe) Tetragonophthalma lives ar- boreally and does not build webs. Specimens examined. — CONGO: 6 paratype of T. balsaci, (ES 8536) [ex syntype of T. vulpina] (MNHN). CAMEROON: 9 holotype of T. crassa (Sjostedt leg. 1891) (NHRS 1403). EQUATORI- AL GUINEA: Benito-River, 9 holotype of T.ferox (BMNH 1898.5.5 101-102) (part). Blandin (1976a: 592) noted three syntypes of ferox; Pocock (1899: 863) described 9(9 holotype in BMNH). TOGO: 9 holotype of T guentheri [Parts of 9 holotype mounted on microscope slides, vial contains re- maining parts of holotype and another adult 9] (ZMHB 13832). IVORY COAST: Lamto, 9 ho- lotype of T. lecordieri (ENS) (MNHN); Lamto, 9 paratype [cited in Blandin (1976a) with incorrect collection number] (MRAC 134.609); Bingerville, S paratype (MRAC 123.720). ZAIRE: Luebo, 9 [marked on vial as paratype, not listed as paratype in Blandin 1976a] (MRAC 12.668). ISL. S. THO- ME: Ribera Palma, 9 holotype of T. thomensis (MCSN). ZAIRE: Gorges de la Pelenge, Upemba National Park, 9 holotype of T. pelengea (MRAC 119.705); 2 d paratypes (MRAC 119.706). T phyl- la (det. Pocock): SIERRA LEONE: Id 19 (BMNH 1898.5.5.95-100). CONGO: 9 holotype of T. vulpina, (ES. 8536) (MNHN). GABON: Id (AMNH). ZAIRE: Mabwe, Upemba National Park, d holotype of T. wittei (MRAC 119.707); 9 para- type (MRAC 119.708). Perenethis L. Koch 1878 Figs. 31-81 Perenethis L. Koch 1878: 980. Type species, by monotypy, Perenethis venusta L. Koch 1878: 980 (9), Australia, Rockhampton. Blandin (1975a) recognized four African species. For the present study, all available type material of African and Asian species was examined. Here, two African species (P. simoni and P. symmetrica), three Asian spe- cies, {P. dentifasciata, P. fascigera, and P. sindica), and one Australian species (P. ven- usta) are recognized. Perenethis huberti Blan- din 1975 and P. lejeuni Blandin 1975 are con- sidered subjective junior synonyms of P. symmetrica (see Sierwald 1989a). Diagnosis. — ^Two cheliceral teeth {ch 0), AER slightly procurved {ch 2), copulatory duct membranous and saccate {ch 13), form- ing two loops {ch 14), of which the first is wider than the second {ch 15), conductor with narrow base {ch 25), tegulum with basal pro- tuberance {ch 21). Autapomorphic characters: Male palp with ventral tibial apophysis {vta, ch 20); conductor with small mesal hump {ch 26) and slender apical section with smoothly rounded tip {ch 24); carina forming two lateral branches {ch 8). Synapomorphic characters: Two cheliceral teeth at retromargin {ch 0) and tegulum with basal protuberance shared with Polyboea and Maypacius {ch 21). Description of characters. — Eye pattern: AER mostly procurved in varying degrees; eyes rather small and subequal, PLE = PME > AME > ALE, PME:AME = 1.2; AME: ALE = 1.2. Chelicerae: Posterior margin with two unequally- sized teeth close to the inner part of the chelicerae. Color pattern: Median yellowish-brown, rarely red-brown; dorsal pattern with light lateral stripes along prosoma and opisthosoma enclose darker median sec- tions; ventral pattern with grayish coloration of legs (especially femora), dark spots on cox- ae, grayish patches on sternum. Spine pattern: Legs as in Charminus, some specimens with thin, short pro- and retrolateral spines at the patella. Palpal femora with thin ventral spines, feature unusual in the Pisaurinae. Spine length 388 THE JOURNAL OF ARACHNOLOGY Figures 31-33. — Perenethis simoni from Ivory Coast (MNHN). 31, Epigynum; 32, Vulva; 33, Left spermatheca, dorsal view. Scale line: 31, 32 = 0.5 mm; 33 = 0.2 mm. [ventral tibial spine, second pair, first leg]: Spine lengthitibia width = 3. Epigynum (Fig. 31): Epigynal folds V-shaped, carina ridge- like, straight (except in P. symmetrica), form- ing two separate lateral branches; fossae lo- cated in the most lateral comers of the carina branches, lateral in relation to copulatory openings. Vulva (Fig. 32): Copulatory duct saccate and membranous forming two large loops, head of spermatheca bent except in P. dentifasciata, spermathecal duct with 3“6 loops, base of spermatheca with small or large lumen. Male palp (Figs. 54-81): Tibia with ventral apophysis; retrolateral tibial apophysis simple and flat, of various lengths, directed forward; tegulum with distinct basal protuber- ance; weakly sclerotized conductor with nar- row base, distinct mesal hump (especially when inflated) and slender apical section with smoothly rounded tip; median apophysis with sclerotized hook; distal tegular apophysis with wing; sclerite A small, oval, elongated; distal sclerotized tube of embolic division short and small, consisting mainly of base; embolus long and whip-shaped, pars pendula about % of embolus length (except P. symmetrica). Natural history. — Koh (1989) collected P. venusta in Singapore in “grassy areas”; label indicates collecting with sweep-net. Blandin (1975a) collected P. simoni among herba- ceous plants. Web-building unknown. REVISION OF THE GENUS PERENETHIS Perenethis dentifasciata (O. Pickard- Cambridge 1885) Figs. 48-50 Ocyale dentifasciata O. Pickard-Cambridge 1885: 79; female holotype; type locality: North-east PAKISTAN or north-west INDIA (“Murree to Sind valley, and Sind valley”); OXUM; vidi. Pisaura dentifasciata, -Simon 1898a: 289. Perenethis dentifasciata, -Sierwald 1987a: 97. Catalogs: Roewer 1954, 2a: 121, sub Pisaura. Bon- net 1955, 2: 3674, sub Pisaura. Platnick 1993: 520, sub Perenethis. Diagnosis. — Carina branches very short, unique within Perenethis (Fig. 48). Description (only $ holotype known).— Body, legs and palps light yellowish-brown; color pattern faded, most hairs lost; remnants of standard dorsal color pattern similar to P. simoni and P. venusta (Figs. 52, 53). Mea- surements: Body 10.6 long, prosoma 3.8 long, 3.25 wide. Leg length: Fe 4.66, PaTi 6.33, MeTa 6.75, total length 17.74. Epigynum (Fig. 48): Central transverse section of carina com- pletely reduced, carina ridges only present around lateral epigynal pits. Vulva (Figs. 49, SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 389 Figures 34, 35. — Perenethis symmetrica from Djibouti (holotype of Perenethis huberti; MNHN), 34, Epigynum; 35, Vulva. Scale line = 0.2 mm. 50): Copulatory duct saccate and membranous with two loops, first loop considerably wider; small head of spermatheca pointing anteriorly; spermathecal duct with three loops; base of spermatheca ball-shaped with large lumen. Male unknown, see under “Special Forms” for a possible male. Distribution.™ Known only from type lo- cality. Perenethis fascigera (Bosenberg & Strand 1906) Tetragonophthalma fascigera Bosenberg & Strand 1906: 306; female holotype; type locality: Japan; Naturkunde-Museum Stuttgart; non vidi (holo- type lost). Perenethis fascigera, -Hu 1984: 260. Yaginuma 1986: 173 ((3 9). Song 1987: 209. Chikuni 1989: Figures 36-38. — Perenethis sindica from Nepal (CM 267, Mechi District, Taplejung). 36, Epigyn- um; 37, Vulva; 38, Left spermatheca, dorsal view. Scale line: 36, 37 = 0.5 mm; 38 = 0.2 mm. 106. Chen & Gao 1990: 136. Chen & Zhang 1991: 225. Catalogs: Roewer 1954, 2a: 118. Bonnet 1955, 2: 4360, sub Tetragonophthalma. Platnick 1989: 394, sub Perenethis', Platnick 1993: 520. Description. — Single S and single $ from Japan). Female: Body, legs and palps yellowish-brown, dorsal color pattern as in P. venusta (Fig. 53). Measurements: Body 390 THE JOURNAL OF ARACHNOLOGY Figures 39-41. — Perenethis sindica from Sri Lanka (OXUM bottle 1526). 39, Epigynym; 40, Vulva; 41, Left spermatheca, dorsal view. Scale line: 39, 40 = 0.5 mm; 41 = 0.2 mm. 8.53 long, prosoma 3.41 long, 2.83 wide; legs all broken off. Epigynum: As in P. sin- dica (Figs. 36, 39). Vulva: As in P. venusta (Figs. 43, 46), spermathecal duct less con- voluted than in P. venusta. Male: Coloration as in female, legs ventrally not dark-gray as in P. venusta but yellowish-brown. Mea- surements: Body 8.33 long, prosoma 3.32 long, 2.6 wide. Leg length: Fe 5.34, PaTi Figures 42-44. — Perenethis venusta from Aus- tralia (lectotype, ZMUH). 42, Epigynum; 43, Vul- va; 44, Left spermatheca, dorsal view. Scale line: 42, 43 = 0.5 mm; 44 = 0.2 mm. 6.41, MeTa 8.34, total length 20.08. Leg for- mula: (I, II) IV, III, leg length differences small. Male palp very similar to P. venusta (Figs. 57), tibial apophysis short, distal teg- ular apophysis with wing. Remarks.— fascigera may be conspecific with P. venusta. Distribution. — Known from Japan and China. Specimens examined. — JAPAN: Kyushu, Ushi- buka, 29 July 1978 (Y. Chikuni). Loan: Courtesy of T. Yaginuma. SffiRWALD— PHYLOGENY OF PISAURINE SPIDERS 391 Figures 45-47. — Perenethis venusta from Aus- tralia (paralectotype; MCSN). 45, Epigynum; 46, Vulva; 47, Left spermatheca, dorsal view. Scale line: 45, 46 = 0.5 mm; 47 = 0.2 mm. Perenethis simoni (Lessert 1916) Figs. 31-33, 51, 52, 54-56) ? Tetragonophthalma phylla, -Simon 1898: 293, CONGO: Landana MNHN ES no. 3080; non vidi (listed by Blandin 1975a: 379). Tetragonophthalma simoni Lessert 1916: 577; 2d, 19 syntypes; type locality: 9 lectotype [here designated], KENYA: Nanyuki [specimen label: Ngare na nyuki]. Expedition Sjostedt; NHRS; d paralectotype [right palp missing], juvenile d paralectotype, TANZANIA: Arusha, Kibonoto [presumably Kibongoto], Expedition Sjostedt; NHRS; vidi. Maypacius berlandi Roewer 1955: 160, nomen no- vum; d 9 syntypes; type locality: ETHIOPIA: Barko; MNHN (det. by Berland as Tetragonoph- thalma stuhlmanni); non vidi. Synonymy by Blandin 1975a: 379. Perenethis straeleni Roewer 1955: 265, d holotype, type locality: ZAIRE, Upemba National Park, Mabwe, Lac Upemba; MRAC 119709; non vidi. Synonymy by Blandin 1975a. Perenethis simoni, -Blandin 1975a: 379; 3d39; IVORY COAST, Lamto, MNHN; vidi. Pisaurellus badicus, -Blandin 1976b: 926, figs. 2, 7b, 8 [non Pisaurellus badicus Roewer 1961, see Perenethis symmetrica below]. Catalogs: Roewer 1954, 2a: 118. Bonnet 1955, 2: 4361, sub Tetragonophthalma. Platnick 1993: 520. Note: In the original description, Lessert (1916: 580) mentioned three specimens (col- lected in Kibonoto [2d, not indicated as sub- adult] and Ngare na nyuki [19] during the Sjostedt Expedition [1905-1906]). One adult male, one subadult male (labelled Kibonoto) and one female (labelled Ngare na nyuki) are deposited in the Naturhistoriska Rijksmuseet in Stockholm. Female here designated as lec- totype. A left male palp (from Kibonoto, MNHNG), labelled syntype, is not part of the adult male syntype from Stockholm (palp is too large). Blandin (1975a: 379) erroneously cited a female in MNHNG as holotype of T. simoni [specimen label states: ZAIRE: Gar- amba]. This female from Zaire, Garamba, was collected during the American Museum Con- go-Expedition in 1937. Diagnosis. — Epigynum with straight carina (ch 9); head of spermatheca bent dorsally pointing anteriorly (ch 16), first loop of sper- mathecal duct forming a complete circle; male conductor with fringed edge. P. simoni very similar to P. sindica and P. venusta, the latter two with fewer loops in the spermathecal duct. Coloration, structure of male and female copulatory organs similar to P. venusta and P. sindica. Description. — Female: (7 9). General coloration of body, legs and palps yellow- ish-brown, prosoma with whitish lateral bands (Fig. 51); opisthosoma with light me- dian band and two whitish stripes laterally; sternum with two dark lateral patches (Fig. 52). Legs ventrally grayish-black, especially the femora. Measurements (9 lectotype): Body 10.9 long, prosoma 4.0 long, 3 wide. Largest female measured: Body 16 long. 392 THE JOURNAL OF ARACHNOLOGY Figures 48-53. — Perenethis, female organs and color pattern. 48-50, P. dentifasciata from Pakistan or India? (Yarkand Mission, holotype; OXUM). 48, Epigynum; 49, Vulva, 50, Left spermatheca, dorsal view. 51-52. — P. simoni from Kenya (lectotype; NHRS): 51, Color pattern sternum; 52, Dorsal color pattern, female. 53, P. venusta from Australia, dorsal color pattern (female lectotype; ZMUH). Scale lines: 48, 49 = 0.5 mm; 50 = 0.1 mm; 51 = 1 mm; 52, 53 = 2 nun. prosoma 4.4 long, 3.3 wide. Leg length (pro- soma 4.0 long): Fe 6.4, PaTi 7.8, MeTa 9.166, total length 23.4. Epigynum (Fig. 31): Straight ridge-like carina conspicuous. Vulva (Fig. 32): Copulatory duct saccate and membranous, forming two large loops, first loop slightly wider than second; head of spermatheca ball-shaped (Fig. 33), bent SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 393 Figures 54-58. — Left male palp of Perenethis. 54-56, P. simoni from Ivory Coast (MNHN). 54, Unex- panded, ventral view; 55, Unexpanded, retrolateral view; 56, Expanded, retrolateral view. 57, P. venusta from Australia (MCSN), unexpanded, retrolateral view. 58, P. symmetrica from South Africa (AMNH), unexpanded, retrolateral view. Scale lines: 54-57 = 1 mm; 58 = 0.5 mm. dorsally; stalk of spermathecae large and strongly sclerotized; spermathecal duct with five loops, the first loop describing a com- plete, small circle, the fourth loop tilted dor- sally; base of spermatheca with small lu- men. Male: (5d). Coloration and pattern as in females. Measurements: Adult paralecto- type: Body 15.16 long, prosoma 5.1 long, 3.75 wide. Smallest specimen: Body 11.2 long, prosoma 4.4 long, 3.5 wide. Leg length (prosoma 4.1 long): Fe 7.0, PaTi 9.08, MeTa 11.6, total length 27.7. Male palp (Figs. 54-56): Retrolateral tibial apophysis long and flat (spatula-shaped), tip rounded; short hump-shaped ventral tibial apophysis, forming a projection of the apical tibial edge; median apophysis narrow, with terminal, sclerotized hook; distal tegular apophysis with terminal hook and conspic- uous “wing”; conductor narrow, partly ex- 394 THE JOURNAL OF ARACHNOLOGY Figures 59-81, — Retrolateral tibial apophysis (odd numbers) and distal tegular apophysis (even num- bers) of left male palp. 59-74, Perenethis sindica. 59, 60, From India: Bombay (OXUM bottle 1522); 61, 62, From India: Bombay (OXUM bottle 1525); 63, 64, From India: Kanchrapara (AMNH); 65, 66, From Sri Lanka (OXUM bottle 1526, tube A); 67, 68, From Sri Lanka (OXUM bottle 1526, tube B); 69, 70, From Nepal: Bagmati Dist., Kathmandu- Valley, Balaju Park, September 1969 (CM); 71, 72, From East Pakistan or northwest India (Yarkand Mission; OXUM). 73, 74, P. simoni from Ivory Coast (MNHN). 75, 76, P. venusta from Australia (note different scale line; MCSN), 77 — 80, Perenethis sp., special forms. 77, 78, From Turkey: Smyrna (OXUM); 79, 80, From Nepal: Dhading Dist, between Kagune and Samari Banjyang, 800-1000 m, 23 July 83, agricultural area (CM). 81, F. symmetrica from South Africa (AMNH), distal tegular apophysis. Scale lines = 0.2 mm. SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 395 pandable, with fringed edge; embolus long, whiplike with pars pendula following 36 of embolus length. Leg formula (d$): (LII), IV, III. Natural history. — Occurs in savanna veg- etation (Blandin 1975a: 380). Distribution. — Africa, south of the Sahara. Specimens examined. — ^Types listed above. TAN- ZANIA: Arusha, Kibonoto, Expedition Sjostedt, Id (left palp only), MNHNG [labelled syntype]. ZAIRE: Garamba (det. Lessert), American Museum Congo- Expedition, 9 labelled holotype by Blandin (MNHNG). IVORY COAST: Lamto, 3d; 3 9 (MNHN). SENEGAL: Dakar, km 15 R. Rufisque, 19, August 1980 (W. Settle) (CASC). BOTSWANA: Serowe, 19, ex malaise trap, March 1990 (P. Forch- hammer) (CASC). ZIMBABWE: 33 mi SE of Chi- rundu, 1170 m elev,. Id, 8 March 1958 (S. Ross & R.E. Leech) (CASC). Perenethis sindica (Simon 1897) Fig. 36--41, 59-72 Tetragonophthalma sindica Simon 1897: 295; 29 syntypes; type locality: INDIA [near Bombay] Kurrachee (MNHN), vidi. Perenethis indica [sic!], -Pocock 1900: 246; 9 (BMNH 99.11.2.147), vidi. Catalogs: Roewer 1954, 2a: 118. Bonnet 1955, 2: 4361, sub Tetragonophthalma. Diagnosis.— -Female copulatory organ very similar to P. simoni and P. venusta, sperma- thecal duct with fewer loops than P. simoni. Male retrolateral tibial apophysis often point- ed {ch 19). Description. — Chelicerae: Inner tooth at posterior margin twice as large at outer tooth. Female: (17 9). Overall coloration yellowish- brown to medium brown; prosoma and opis- thosoma dorsally with broad, dark, median band; set off by narrow, straight stripes of sil- ver or white; sternum and opisthosoma ventrally with pale median band. Opisthosoma slender and elongated. Legs uniformly brown. Measurements: Range: Body 8.7 long, pro- soma 3.3 long, 2.8 wide (Sri Lanka) to body 20.4 long, prosoma 5.8 long, 4.5 wide (IN- DIA: West Bengal). Leg length (prosoma 3.75 long): Fe 5.7, PaTi 7.1, MeTa 8.6, total length 21.5. Epigynum (Figs. 36, 39): Very similar to P. simoni and P. venusta, carina variable. Vulva (Figs. 37, 38, 40, 41): Copulatory duct and spermatheca very similar to P. simoni and P. venusta; spermathecal duct with four loops; head of spermatheca bent as in P. simoni and P. venusta; lumen of base of spermatheca large as in P. venusta and larger than in P. simoni. Male: (10<3). Shape, color and color pattern of body and legs as in female. Mea- surements: Range: Body 8.75 long, prosoma 3.5 long, 2.9 wide (Sri Lanka) to body 16.4 long, prosoma 5.6 long, 4.1 wide (INDIA: Bengal). Leg length (prosoma 3.75 long): Fe 7.08, PaTi 9.16, MeTa 11.1, total length 27.4. Male palp (Figs. 59-72): Palp and genital bulb similar to P. simoni and P. venusta; dis- tinct ventral tibial apophysis with two- pronged tip, larger than in P. simoni; retrola- teral tibial apophysis flat and spatula-shaped as in P. simoni, but with variations in length and shape, tip pointed. Great variation in dis- tal tegular apophyses, especially in form of hook and wing. Remarks. — The size range for the speci- mens appears to be very high. In addition, fea- tures of the male copulatory organ are sur- prisingly variable, but disjunct concordant features can not be found in the sample avail- able for this study (Material from the National Collection in Calcutta was not available for study, Biswas in litt. 1987). Natural history. — No data are available. Distribution.— India, Sri Lanka, Philip- pines. Specimens examined. — INDIA: SE West Ben- gal, Kanchrapara, 1 c329 , July 1944 (AMNH); same locality, Id (AMNH); Maharashtra, Bombay, Id, (OXUM 1525); same locality. Id, (OXUM 1522, tube 90); Maharashtra, Pune, 1 9 (BMNH 1899.11.2.147). SRI LANKA: 4d6 9 (OXUM 1526); 19 (ZMHB 29225). NEPAL: Dhang Dist. W Samari, Banjyang/Topal Khola (river), agricul- tural area, 1000-1200 m, 1 9 , 23 July 83 (Martens & Schawaller leg.) (CM 211); Mechi District, Ta- plejung, Kabeli Khola (river), 900-1250 m, agri- cultural area, forest remains, 19,1 September 83 (Martens & Daams leg.) (CM 267); Bagmati Dist., Kathmandu- Valley, Balaju Park, 1 d 1 9 , September 69 (CM without number). PHILIPPINES: Luzon, 1 9 (ZMHB 3847). Locality unknown, “Yarkand- Mission”, Id 19 (OXUM). Perenethis symmetrica (Lawrence 1927) Figs. 34, 35, 58, 81 Tetragonophthalma symmetrica Lawrence 1927: 46; female holotype; type locality: NAMIBIA: Ongandjera; SAM B 6228; vidi. Perenethis symmetrica, -Roewer 1955: 267. Pisaurellus badicus Roewer 1961: 40, fig. 5; 9 ho- lotype; type locality: SENEGAL: Parc National 396 THE JOURNAL OF ARACHNOLOGY du Niololo-Koba; non vidi. Holotype not avail- able from IFAN. NEW SYNONYMY. Perenethis huberti Blandin 1975a: 382; 9 holotype, 49 and 2 juv. paratypes, type locality: AFAR: Djibouti, MNHN No. 19149; vidi. Synonymy by Sierwald 1989a. Perenethis lejeunei Blandin 1975a: 382; 9 holo- type, 6 (= paratype); type locality: ZAIRE: Kivu; MRAC 144355; non vidi. Synonymy by Sierwald 1989a. Catalogs: Roewer 1954, 2a: 118. Bonnet 1955, 2: 4361, sub Tetragonophthalma. Brignoli 1983: 463, P. huberti’, page 464, P. lejeunei. Platnick 1993: 520. Taxonomic note: The specimen figured by Blandin (1976b: fig. 8, male palp) as Pisau- rellus badicus is not conspecific with P. bad- icus Roewer 1961, but belongs to Perenethis simoni. Roewer’s figures (1961; fig. 5 c,e,d) of the palp establish the synonymy of Pisau- rellus badicus with Perenethis symmetrica. Diagnosis. — Color pattern distinctive (Blandin 1975a: figs. 8, 10): Opisthosoma dorsally with dark, lobed median band; carina of epigynum curved like eyebrows {ch 9); both loops of membranous copulatory duct of approximately same size {ch 15); embolus dis- tinctive, broad pars pendula reaching tip of embolus {ch 34); smallest species of the ge- nus. Description. — Chelicerae: Inner tooth at posterior margin only slightly larger than out- er tooth. Leg formula: (I-II), IV, III. Very little variation in color pattern and copulatory or- gans. Female: (119). Prosoma: Dorsally with dark median band, thin bright line in the mid- dle, and laterally two broad light-yellow lon- gitudinal bands; sternum yellow with three dark-gray spots at each side; opisthosoma dorsally with dark, lobed median band (straight in all other species of Perenethis)’, sides yellowish with irregular brown mark- ings; venter yellowish without pattern. Legs: Femora to tibia of first three legs ventrally dark gray. Palps with black rings at joints. Measurements: Type P. symmetrica’. Body 8.5 long, prosoma 2.5 long, 2.0 wide. Range: Body 5.7 long, prosoma 2.1 long, 1.8 wide to body 8.5 long, prosoma 2.5 long, 2.0 wide. Leg length (prosoma 2.3 long): Fe 3.1, PaTi 3.8, MeTa 4.0, total length 11.0. Epigynum (Fig. 34): Carina curved, forming “eye- brows.” Vulva (Fig. 35): Copulatory duct membranous, in two large, nearly equal-sized loops; spermatheca large, head of spermatheca round, pointing laterally; stalk thick, sperma- thecal duct with seven loops; base of sper- matheca with small lumen. Male: (15 d). Shape, color and color pattern of body and legs similar to female. Legs ventrally lighter gray, only light markings at palps. Measure- ments: Body 5.3 long, prosoma 2.3 long, 1.7 wide to body 7.2 long, prosoma 3.0 long, 2.7 wide. Leg length (prosoma 2.3 long): Fe 4.2, PaTi 5.3, MeTa 6.2, total length 15.7. Male palp (Figs. 58, 81): Ventral tibial apophysis distinct with swollen tip; retrolateral tibial apophysis spatula-shaped, with rounded tip, little variation in the examined sample; con- ductor slender, edge without fringe; median apophysis with rather large, sclerotized hook; distal tegular apophysis with small hook, without wing; embolus moderately long, pars pendula nearly reaching tip of embolus. Natural history. — Occurs in shrubs (Blan- din 1975a). Distribution. — Africa, south of the Sahara, reaching well into South Africa. Specimens examined. — Types listed above. SOUTH AFRICA: Transvaal, Kruger Park, Sku- kuza, in thomscrub, 126, 15 December 1985 (AMNH); east Transvaal, 15 km from Klaserie, Guernsey Farm, woodland, 5c?494juv9, 19-31 December 1985 (AMNH). TANZANIA: Tabora, \6 (ZMHB 29226). KENYA: Lake Nakuru Na~ tional Preserve, campsite in yellow fever forest, 5(34 9, 14 May 1975 (A.J. Penniman leg.) (AMNH). ZAIRE: Epulu, 250 m, lc3, 2 October 57 (E.S. Ross & R.E. Leech) (CASC). Perenethis venusta L. Koch 1878 Figs. 42-47, 53, 57, 75, 76 Perenethis venusta L. Koch 1878: 980; 3 9 synty- pes; type locality: AUSTRALIA, Queensland. 9 lectotype here designated; ZMUH, vidi. 9 para- lectotype: Rockhampton; ZMHB 3501 (opistho- soma missing); vidi. 9 paralectotype: Peak Down; BMNH; vidi. Perenethis venusta, -Thorell 1881: 372; (39 spec- imens, locality: AUSTRALIA: Queensland, Cape York Peninsula; MCSN; vidi. Perenethis unifasciata, -Thorell 1891: 61; P. ven- usta placed in synonymy of P. unifasciata. Perenethis parkinsoni Dahl 1908: 228; 9 holotype, type locality: PAPUA-NEW GUINEA: Bismark- Archipelago, Ralum; ZMHB 29 224; vidi. Spec- imen demounted from microscope slide, NEW SYNONYMY Perenethis unifasciata, -Chrysanthus 1967: 421; several 6 9 specimens from INDONESIA: New SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 397 Guinea, Merauke and Mindiptana; non vidi, fig- ures agree with P. venusta. P. venusta, -Chrysanthus 1967: 421, fig. 58, figure probably based on a subadult Perenethis female; removed P. venusta from synonymy of P. uni- fas data. Perenethis venusta, -Todd-Davies 1985: 104. Catalogs: Roewer 1954, 2a: 118, as synonym of P. unifasdata. Bonnet 1955, 2; 4361, sub Tetragon- ophthalma, as synonym of P. unifasdata. Plat- nick 1993: 520. Taxonomic note. — P. parkinsoni is based on a single female specimen. The female cop- ulatory organ is very similar to Thorell’s ven- usta specimen; (Figs. 45-47). Diagnosis.— -Color pattern and copulatory organs very similar to P. simoni and P. sin- dica, spermathecal duct with fewer loops than P. simoni. Description.— Chelicerae: Inner tooth of both teeth at posterior margin distinctly larger than outer tooth. Leg formula: (II-I), IV, III. Female: (13$). Coloration light yellowish- brown. Many specimens with dark-grayish coloration on the ventral side of femora, dark gray spots on the coxae and two dark-gray patches on the sternum as in P. simoni (Fig. 51). Dorsal color-pattern (Fig. 53) very con- sistent, prosoma with dark median band and light lateral zones. Two thin stripes of white hairs separate median zone from lateral zones. Opisthosoma with straight dark median band, two thin stripes of dark hair separating the median band from the light-colored lateral zones. Ventral color-pattern: Light median band caused by guanine, laterally two thin dark bands followed by two white bands formed by hair. Lateral parts of opisthosoma covered with grayish-brown hair. Measure- ments lectotype (ZMUH): Body 10.4 long, prosoma 3.9 long, 2.9 wide. Females slightly smaller than males, legs shorter. Range [13 $]: Body 7.7 long, prosoma 2.87 long, 2.25 wide; to body ca. 13 long, prosoma 4.5 long, 3.2 wide. Leg length (prosoma 4.14 long): Fe 5.96, PaTi 7.63, MeTa 8.36, total length 21.95. Epigynum in two rather distinct forms (Figs. 42, 45) both equally common. Carina branch- es either nearly adjoining in the middle or dis- tinctly separated; external copulatory opening rather large. Vulva (Figs. 43, 44, 46, 47): Cop- ulatory duct membranous, wide and saccate, forming two loops, second loop much narrow- er than first. Small head of spermatheca and adjacent slender stalk of spermatheca bent dorsally; this part of the spermatheca is small- er than in P. sindica. Remaining spermatheca thick and heavily sclerotized; spermathecal duct either with three or four loops, loops slightly variable; size of lumen of base of spermatheca rather large but variable. Variable features of vulva not correlated with either epigynum- type. Female copulatory organ very similar to P. sindica and P. simoni. Male: {IS). Coloration and color pattern as in fe- males, somewhat lighter. Measurements [7d]: Males slightly larger than females with longer legs; body 10.6 long, prosoma 4.0 long, 2.8 wide to body 12.0 long, prosoma 4.72 long, 3.56 wide. Leg length (prosoma 4.07 long): Fe 6.98, PaTi 9.16, MeTa 10.61, total length 26.76. Male palp (Figs. 57, 75, 76): Very sim- ilar to P. simoni and P. sindica', retrolateral tibial apophysis (Fig. 57) long and flat (spat- ula-shaped), tip bluntly pointed; tibial apical margin with low projection similar to P. si- moni; median apophysis narrow, with termi- nal, sclerotized hook, membranous base of median apophysis enlarged; distal tegular apophysis with terminal hook and conspicu- ous “wing” (Fig. 76); conductor genus-typi- cal, slender, without fringe; embolus long, whip-like with conspicuous pars pendula. Form of retrolateral tibial apophysis, median apophysis and distal tegular apophysis show very little variation within the Australian specimens; the male from Singapore very similar as well. Natural history. — Occurs in grassland and forests (specimen labels; QMBA); Koh (1989). Distribution.— Thailand, Singapore, Aus- tralia and Papua New Guinea. Specimens examined. — AUSTRALIA: Queens- land: Homeval, northeast Qld, 1$ (QMBA S14 644). Eureka Ck, 1 $, 2 Febmary 72 (QMBA S 14 634). Rundle Ra, northeast QLD, 1 $ , 31 March 75 (QMBA S14 648). Doboy Ck, southeast QLD, 131$, 9 January 79 (QMBA S14 632). Brisbane, 1$ with egg sac, 16 March 86 (QMBA S14 630). Bald Hills, southeast QLD, 1 $ , 20 December 79 (QMBA S14 636). Bald Hills, 13, 10 January 80 (QMBA S14 639). Cape Hillsborough, N.R grass area, 1$, 5 January 75 (QMBA S14 643). Bunda- berg forest, southeast QLD, 1 $ (QMBA S14 629). Newroy Is. N.R, 13, 14 Febmary 75 (QMBA S14 638). Curmmbin, southeast QLD, 13, 11 January 80 (QMBA S14 633). Weipa, 13, 7 Febmary 75 (QMBA S14 641). 12 samples with juveniles from 398 THE JOURNAL OF ARACHNOLOGY Queensland Museum. SINGAPORE: Mac Ritchi Reservoir, in grass, 1$ (Koh 89.07.13.08). Mal- colm Road, grassy waste land, 1 S , (Koh 85.08.24.01). PAPUA NEW GUINEA: Madang Province, Sapi Forest Reserve, 30 km west of Ma- dang, 5°12'S, 145°30E, 19,4 July 1988 (W.J. Pu- lawski) (CASC). Vogelkop, Manokwan, 75 m, 19, 21 July 1951 (D. Elmo Hardy) (BPBM). Waris, 450 m, 1 9 , VII- VIII (TC. Maa) (BPBM). THAILAND: 8 mi SE Saraburi, 100 m, 1 (3, 28 July 62 (E.S. Ross & D.Q. Cavagnaro) (CASC). Special forms. — The material examined for this study contained two males that cannot be placed in any described Perenethis species. Due to the uncertainty of species discrimina- tion between P. fascigera, P. simoni, P. sin- dica and P. venusta on one hand, and the un- usual variability in P, sindica on the other hand, descriptions of new species-group taxa do not appear justified at this point. Form I (Figs. 77, 78): Male from Turkey: Smyrna, OXUM. The specimen is similar to P. simoni, but the retrolateral tibial apophysis possesses an anterior basal projection. Form II (Figs. 79, 80): Male from Nepal [Dhading Dist, between Kagune and Samari Banjyang, 800-1000 m, 23 July 83, agricultural area, CM]. In this specimen the shape of the retro- lateral apophysis is different and does not fit the overall pattern found in P. sindica. Since the shape of the retrolateral tibial apophysis is often species-typical in Pisauridae, this spec- imen could belong to a species distinct from P. sindica; it may represent the male of P. dentifasciata. NOMINA DUBIA Perenethis brevipes (Strand 1906) Tetragonophthalma brevipes Strand 1906: 685; ho- lotype juvenile (lost), type locality: Sudan, Har- erge Mountains; Naturkunde-Museum Stuttgart. Perenethis brevipes, -Roewer 1955: 267. P. brevP pes, -Blandin 1975a: 384; nomen dubium Ctenus marginatus Walckenaer 1847 Ctenus marginatus Walckenaer 1847: 402, 9 ?ho- lotype; type locality; Solomon Islands; type pre- sumed lost. Thalassius marginatus, -Simon 1891: 299. Walckenaer compares the specimen to Pi- saura mirabilis. This could suggest that Wal- ckenaer’s specimen was congeneric with Per- enethis (general color pattern and habitus). Simon’s (1891) placement of this species in the genus Thalassius was rejected (Sierwald 1987). Perenethis rectifasciata (O. Pickard- Cambridge 1885) Ocyale rectifasciata O. Pickard-Cambridge 1885: 78; juvenile male holotype; type locality: north- east PAKISTAN or north-west INDIA (“Murree to Sind valley and Sind valley”); OXUM; vidi. Pisaura rectifasciata, -Simon 1898a: 289. Catalogs: Roewer 1954, 2a: 121, sub Pisaura. Bon- net 1955, 2: 3681, sub Pisaura. Based on eye-pattem and number of teeth at the chelicerae, the subadult male is a mem- ber of the genus Perenethis. Color-pattern fad- ed, most spines lost. P. rectifasciata is here considered a nomen dubium. Perenethis unifasciata (Doleschall 1859) Dolomedes unifasciata Doleschall 1859: 10; 9 ho- lotype lost; type locality: Indonesia: Amboina. Perenethis unifasciata, -Thorell 1891; 61. Tetragonophthalma unifasciata, -Strand 1911: 165, 9 from INDONESIA: Kepulauan Am Islands, Pulau Kobroor. Specimen not in SMFD (fide Chrysanthus 1967). Catalogs: Roewer 1954, 2a: 118. Bonnet 1955, 2: 4361, sub Tetragonophthalma. According to the collection catalog in the Museum for Natuurlijke Historic in Leiden the female specimen figured in DoleschalFs publication (1859, fig. 6) never arrived in Lei- den (van der Hammen pers. comm. 1982). Therefore, no actual type-specimen exists. The specimen figured could be conspecific with venusta. P. unifasciata is here considered a nomen dubium. Maypacius Simon 1898 Figs. 82-87, 91-96 Maypacius Simon 1898a: 292. Type species, by original designation, Maypacius vittiger Simon 1898b: 13; female holotype, Madagascar & Af- rica. Blandin (1975a, 1978b) recognized nine species in the genus Maypacius and listed a total of 21 specimens, only six of them males; five species are known from females only, two from males; for two species both sexes axe recognized. Species: Tetragonophthalma bill- neatus Pavesi 1895, 9 known; Maypacius christophei Blandin 1975 (1975a), $ known; Maypacius curiosus Blandin 1975 (1975a), S known; Maypacius gilloni Blandin 1978 (1978b), d 9 known; Maypacius kaestneri SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 399 Roewer 1955, S 9 known; Maypacius pe~ trunkevitchi Lessert 1933, 9 known; Maypa- cius roeweri Blandin 1975 (1975a), S known; Maypacius stuhlmanni Bosenberg & Lenz 1894, 9 known; Maypacius vittiger Simon 1898 [Simon 1898b: 13], 9 known. Maypa- cius vittiger was synonymized with Tetragon- ophthalma bilineatus Pavesi 1895 by Simon (1906: 1169); Roewer (1955: 153) listed M. vittiger as junior synonym of Maypacius bil- ineatus', Blandin (1974a: 309; 1975a: 385) re- moved M. vittiger from the synonymy of M. bilineatus. Diagnosis. — Two equally-sized cheliceral teeth {ch 0, 1), short copulatory duct {ch 15), short spines {ch 6), and the following auta- pomorphic characters: Strongly procurved AER {ch 2), ALE on tubercles and located nearly beneath AME; conductor short {ch 23), with specialized apical region with two guid- ing lamellae {ch 28). Synapomorphic charac- ters: Two cheliceral teeth at retromargin {ch 0) and retrolateral peak at tegulum {ch 21) shared with Polyboea and Perenethis', con- ductor with two guiding lamella {ch 28), pit in dorsal branch of distal tegular apophysis {ch 29) and shape of sclerite A shared with Polyboea {ch 31). Description. — Based on M. kaestneri, M. petrunkevitchi, and M. roeweri. Measure- ments: M. kaestneri: 9 : Body 12.43 long, pro- soma 3.56 long, 2.6 wide (MRAC 142.407). Leg length: Fe 6.6, PaTi 7.8, MeTa 8.7, total length 23.2. M. petrunkevitchi: 9 : Body 13.45 long, prosoma 2.9 long, 2.18 wide. Leg length: Fe 5.45, PaTi 6.6, MeTa 7.2, total length 19.25 (MRAC 145.395). M. roeweri: $: Body 11.81 long, prosoma 3.45 long, 2.7 wide. All legs broken off. Eye pattern: AE in two rows (AER extremely procurved); ALE on tubercles, in front of AME and only slight- ly further apart from each other than AME; PLE=AME>PME”ALE. Eyes small com- pared to body size, PME:AME 0.7-0. 8; AME: ALE = 1,5. Chelicerae: Posterior mar- gin with two equally-sized teeth, teeth closer to outer edge of chelicerae and wider spaced than in Perenethis. Spine pattern identical with Charminus camerunensis . Spine length: Spines very short; spine length:tibia width = 1. Epigynum (M. petrunkevitchi and M. kae- stneri, Figs. 82, 85): Continuous carina weak- ly developed, anterior edge conspicuous, ca- rina straight or recurved (strongly recurved in Figures 82-84. — Maypacius petrunkevitchi from Rwanda (MRAC 145.395). 82, Epigynum; 83, Vul- va; 84, Left spermatheca, dorsal view. Scale lines: 82, 83 = 0.5 mm; 84 = 0.2 mm. M. petrunkevitchi)', fossae mesal to copulatory opening. Vulva (Figs. 83, 84, 86, 87): Mem- branous copulatory duct short, forming single curve; M. petrunkevitchi: Copulatory duct sclerotized close to the spermathecae, head of spermatheca pointing anteriorly, spermathecal duct forming single loop, base of spermatheca with small lumen; M. kaestneri: Head of sper- matheca bent, spermathecal duct with two loops, base of spermatheca with large lumen. 400 THE JOURNAL OF ARACHNOLOGY Figures 85-87. — Maypacius kaestneri from Gha- na (MRAC 142.407). 85, Epigynum; 86, Vulva; 87, Left spermatheca, dorsal view. Scale lines: 85, 86 = 0.5 mm; 87 = 0.2 mm. Male palp (based on M. roeweri. Figs. 91-96): Retrolateral tibial apophysis pointed, directed forward (perpendicular in M. curiosus); te- gulum with conspicuous, retrolateral peak; short conductor with narrow base and unique- ly enlarged tip, embolus resting between two lamellae; distal tegular apophysis with hook and wing, dorsal branch of distal tegular apophysis with pit as in Polyboea; sclerite A large, forked, similar to Polyboea; distal scler- otized tube similar to Polyboea; embolus Figures 88-90. — Polyboea vulpina from Singa- pore (NMSC). 88, Epigynum; 89, Vulva; 90, Left spermatheca, dorsal view. Scale lines: 88, 89 = 0.5 mm; 90 = 0.2 mm. short, with wide pars pendula, about Vi em- bolus length. For the cladistic analysis, char- acters for the male of M. kaestneri were taken from Blandin’s figure (1975a: 389, fig. 21, 22). Natural history/habitat. — Occurs in the savanna, found in vegetation (Blandin 1978b). Specimens examined. — M. roeweri: ZAIRE: Kivu, Uvira, Mugesera, 1 6 paratype (MRAC 145.058). M. petrunkevitchi: RWANDA: Burge- sera, Biharagu, found in dense field vegetation, 1 9 (MRAC 145.395); Butare, 29 (MRAC 140.720). M. kaestneri: GHANA: Legon, 1 9 (MRAC SEERWALD— PHYLOGENY OF PISAURINE SPIDERS 401 142.407). CONGO: Faradje, 1 $ (MRAC 145.400). Polyboea Thorell 1895 Figs. 88, 90, 97-101 Polyboea Thorell 1895: 228. Type species, by orig- inal designation, Polyboea vulpina Thorell 1895: 229, juvenile male holotype, Burma: Rangoon). The genus is based on a subadult male of the type species from Burma. Male and fe- male copulatory organs from specimens col- lected in Singapore are figured here for the first time. Currently, the genus is monotypic. The Asian pisaurid genus Eurychoera Thorell 1897 (listed in the Pisaurinae by Roewer, 1955: 115) from Singapore (Koh 1989: 97) is not closely related to Polyboea. Diagnosis. — -AER procurved {ch 2), two equally-sized cheliceral teeth {ch 0) and the following autapomorphic characters: ALE sig- nificantly larger than PME {ch 3) and AME {ch 4); chelicerae longer than in all other per- enethine genera; absence of the two paired short spines apically at the ventral side of the tibia. Since the genus is currently monotypic, characters listed here may be apomorphic at species level. Synapomorphic characters: Two cheliceral teeth at retromargin {ch 0) and te- gulum with retrolateral peak {ch 21) shared with Perenethis and Maypacius; copulatory duct with two wide membranous loops shared with Perenethis and Charminus camerunemis {ch 14), conductor with two guiding lamellae {ch 28), pit in dorsal branch of distal tegular apophysis {ch 29) and shape of sclerite A shared with Maypacius {ch 31). Polyboea vulpina Thorell 1895 Figs. 88-90, 97-101 Polyboea vulpina Thorell 1895: 229. Polyboea vulpina, -Workman & Workman 1897: 97 (= Ocyale hirsuta on plate 97.); non vidi. PolybaeaLic] vulpina, -Simon 1898a: 289, 296 Polyboea vulpina, -Hasselt 1899: 174 Polyboea vulpina, -Koh 1989: 100 (color photo of d) Catalogs: Petrankevitch 1928: 102, as Polybaea. Roewer 1954, 2a: 122, as Polybaea. Bonnet 1955, 2: 3751. Genus listed: Brignoli 1983: 461, Polybaea; Platnick 1989: 393, Polybaea; Platnick 1993: 521, Polyboea. Diagnosis.- — -Chelicerae large, spines long, tibia lacks apical ventral spine pair; epigynal folds parallel and apart from each other an- teriorly; Carina forming lip with straight pos- terior edge; long conductor with curved tip, similar to conductor in the west African May- pacius gilloni (Blandin 1978b: fig. 2). Description. — Eye pattern: AER procur- ved; AER nearly as wide as PER, ALE>PLE=PME>AME. AME conspicuous- ly smaller than PME, PME: AME = 1.6; AME:ALE = 0.5. Chelicerae: Posterior mar- gin with two nearly equally- sized teeth, teeth closer to outer edge of chelicerae and wider spaced than in Perenethis; chelicerae in both sexes longer than in other Pisauridae, prosoma width :chelicerae-length = 1.7; compare to Maypacius roeweri S : prosoma width:chelic- era length = 2.7; petrunkevitchi ?: 2.39; kae- stneri $ : 3.3. Spine pattern (Table 6): Pro- and retrolateral femoral spines variable within a single specimen. The absence of two paired short spines apically at the ventral side of the tibia is unique within the Pisaurinae. Spine length: Spines very long; spine length: tibia width = 6.5. Female: (1 $). Light orange-yel- low, pattern faded (but see male coloration be- low). Measurements: Body 6.03 long, proso- ma 2.5 long, 2 wide. Epigynum (Fig. 88): Epigynal folds parallel and apart from each other anteriorly; adjoining posteriorly; carina forming lip with straight posterior edge, over- hanging copulatory opening; fossae close to- gether, mesal to the copulatory openings. Vul- va (Figs. 89, 90): Copulatory duct wide and membranous, forming two saccate loops as in Perenethis, first loop larger than second loop; head of spermatheca bent, pointing anteriorly; spermathecal duct with four loops; base of spermatheca with small lumen. Male: {IS). Carapace, legs and sternum light orange-yel- low, abdomen dorsally with distinct gray- beige Y-shaped figure, the anterior lateral stripes of the Y meet behind the heart, a pair of distinct white spots lateral to the median tail- stripe of the Y. Abdomen ventrally with two parallel narrow dark lines. Measurements: Body length 9.16-10.6, prosoma 3.08 — 3.6 long, 2.36-2.96 wide. Leg length (prosoma 3.6 long): Fe 6.3, PaTi 8.16, MeTa 9.4, total length 23.8. Male palp (Figs, 97-101): Retro- lateral tibial apophysis perpendicular, tip with two pointed ends; tegulum with conspicuous peak at retrolateral comer; long conductor with narrow base and broad tip; tip curved in a spiral; two long guiding lamellae, especially visible in the expanded palp; distal tegular apophysis small, with fringed wing; sclerite A 402 THE JOURNAL OF ARACHNOLOGY Figures 91-101. — Maypacius roeweri and Polyboea vulpina. 91-96, Maypacius roeweri from Zaire (MRAC 145.058). 91, Unexpanded left palp, ventral view; 92, Unexpanded left palp, retrolateral view; 93, Expanded right palp, retrolateral view, pit indicates pit in distal tegular apophysis; 94, Expanded right SffiRWALD— PHYLOGENY OF PISAURINE SPIDERS 403 large with the straight edge visible in the unexpanded palp as in Maypacius, forked dis- tal end towards the distal sclerotized tube; em- bolus moderately long; pars pendula short and wide as in Maypacius. Natural history/habitat. — Hasselt (1899) and Koh (1989) report that P. vulpina occurs in grasses and low shrubs, building “large, three-dimensional webs that may be connect- ed with one another.” This may indicate some form of colonial habit. Distribution. — ^Known from Thailand, Ma- laysia and Singapore. Specimens examined. — SINGAPORE: no lo- cality given, 1$ (NMSC 1990.600). Mac Ritchie Reservoir, grasses. Id (Koh 77.01.01.03). THAI- LAND: Khao. Yai Nat. Park, 750 m, 2d, 26 July 1962 (E.S. Ross & D.Q. Cavagnaro) (CASC). 10 mi N Saraburi, 100 m, 1 juv.d, 11 July 1962 (E.S. Ross & D.Q. Cavagnaro) (CASC). 20 mi S.E. Chantaburi, 75 m, 2d, 1 August 1962 (E.S. Ross & D.Q. Cavagnaro) (CASC). MALAYSIA: Fra- ser’s Hill, 4200 m, 1 d, 2juv., 17 June 62 (E.S. Ross & D.Q. Cavagnaro) (CASC). ACKNOWLEDGMENTS I am grateful to the following colleagues, who made the specimens available to me: Dr. Arbocco, Museo Civico di Storia naturale “G. Doria,” Genova (MCSN), Italy; Dr. J. Cod- dington, Mr. S. Larcher, National Museum of Natural History, Smithsonian Institution, Washington, DC (USNM); Dr. M. Grasshoff, Forschungsinstitut Senckenberg (SMFD), Frankfurt/Main, Germany; Dr. C. Griswold, California Academy of Sciences (CASC), San Francisco; Dr. B. Hauser, Museum Geneve (MHNG), Switzerland; Dr. J. Hertault, Dr. C. Rollard, Museum Nationale d’Histoire Natu- relle (MNHN), Paris, France; Mr. R Hillyard, The Natural History Museum (BMNH), Lon- don, Great Britain; Hope Entomological Col- lections, University Museum, Oxford, UK (OXUM). Dr. R. Jocque, Musee Royal de I’Afrique Centrale (MRAC), Tervuren, Bel- gium; Mr. J. Koh (Coll. Koh), Singapore; Mr. T. Kronestedt, Naturhistoriska Rijkmuseet (NHRS), Stockholm, Sweden; Dr. S. Mahun- ka, Termeszettodomanyi Muzeum, Budapest (HNHM), Hungary; Ms S. Mascherini, Museo Zoologico della Specola (MZUF), Firenze, It- aly; Dr. M. Moritz, Zoologisches Museum der Humboldt-Universitat (ZMHB), Berlin, Ger- many; Dr. N.I. Platnick, American Museum of Natural History (AMNH), New York, USA; Dr. G. Rack, Zoologisches Institut und Mu- seum (ZMUH), Universitat Hamburg, Ger- many; Dr. R. Raven, Queensland Museum (QMBA), Brisbane, Australia; Dr. S. Riechert (Coll. Riechert), Knoxville, Tennessee, USA; Dr. S.E Swift, Bishop Museum, Honolulu, Ha- waii (BPBM); Prof. Dr. T. Yaginuma (Coll. Yaginuma), Osaka, Japan; Zoological Refer- ence Collection (NMSC), Singapore. Research funding was provided in part by a grant of the Deutsche Forschungsgemein- schaft (DFG, Germany). A grant from “The Exline-Frizzell Fund for Arachnological Re- search” supported a visit to the California Academy of Sciences. Harbor Branch Ocean- ographic Institution (Ft. Pierce, Florida) and the Delaware Museum of Natural History (Wilmington, DE) were my host institutions during part of the study. Drs. J. Coddington and C. Griswold granted me the much appre- ciated favor of many cheerful and enlighten- ing discussions concerning Hennig86, the Ly- cosoidea and spider copulatory organs. I wish to thank Drs. Bieler, Coddington, Griswold, and Scharff, and especially two anonymous reviewers for the Journal of Arachnology for constructive criticism and candid comments on earlier drafts of the manuscript. LITERATURE CITED Arnett, R.H., G.A. Samuelson & G.M. Nishida. 1993. The insect and spider collections of the world. Sandhill Crane Press, Gainesville, Flori- da, 310 pp. Axelrod, D.I. & P.H. Raven. 1978. Late Cretaceous and Tertiary vegetation history of Africa. Pp. 77- 130. In Werger, M. J.A. (ed.). Biogeography and palp, ventral view; 95, Unexpanded left palp, embolic division, top view; 96, Expanded right palp, embolic division, lateral view. 97-101, Left male palp of Polyboea vulpina from Singapore (Coll. Koh)., Singapore. 97, Unexpanded, ventral view; 98, Expanded, prolateral view, pit indicates pit in distal tegular apophysis; 99, Same, embolic division, top view; 100, Apophysis distal tegular apophysis; 101, Retrolateral tibial. Scale lines = 0.5 mm. 404 THE JOURNAL OF ARACHNOLOGY ecology of southern Africa. The Hague. Dr. W. Junk Publishers. Berland, L. 1922. Araignees. Pp. 43-90. In Voyage de M. le Baron de Rothschild en Ethiopie et en Afrique orientale anglaise (1904-1905). ResuL tats scientifiques. Animaux articles. Ire Partie, Paris. Blandin, P. 1974a. Etudes sur les Pisauridae Afri- caines. I. Le probleme de la definition des genres Perenethis L. Koch, 1878, Tetragonophthalma Karsch, 1878, Chiasmopes Pavesi, 1883 d May- pacius Simon, 1898. (Araneae -Pisauridae -Pi- saurinae). Rev. Suisse ZooL, 81:305-318. Blandin, P. 1974b. Etudes sur les Pisauridae Afri- caines. II. Definition du genre Euprosthenops Po- cock, 1897 et description du genre Euprosthnop- sis n. gen. (Araneae -Pisauridae -Pisaurinae). Rev. Suisse ZooL, 81:933-947. Blandin, P. 1975a. Etudes sur les Pisauridae Afri- caines. III. Les especes des genres Perenethis L. Koch, 1878 d Maypacius Simon, 1898 (Araneae -Pisauridae -Pisaurinae). Rev. ZooL Africaines, 89:376-393. Blandin, P. 1975b. Note sur Euprosthenops ellioti (Pickard-Cambridge O., 1877) (Araneae -Pisaur- idae -Pisaurinae). Bull. Soc. ZooL France, 100: 575-581. Blandin, P. 1976a. Etudes sur les Pisauridae Afri- caines. V. Mise au point sur les especes du genre Tetragonophthalma Karsch, 1878 (Araneae -Pi- sauridae -Pisaurinae). Rev, ZooL Africaines, 90: 588-606. Blandin, P. 1976b. Etudes sur les Pisauridae Afri- caines. VI. Definition des genres Pisaura Simon, 1885, Pisaurellus Roewer, 1961, Afropisaura n. gen. et mise au point sur les especes des genres Afropisaura et Pisaurellus (Araneae -Pisauridae -Pisaurinae). Rev. ZooL Africaines, 90:917-939. Blandin, P. 1977. Etudes sur les Pisauridae Afri- caines. VIIL Les genres Chiasmopes Pavesi, 1883 et Rothus Simon, 1898 (Araneae -Pisauri- dae -Pisaurinae). Rev. ZooL Africaines, 91:538- 557. Blandin, P. 1978a. Etudes sur les Pisauridae Afri- caines. IX. Mise au point sur les genres Cispius Simon, 1898 et Charminus Thorell, 1899. (Ara- neae -Pisauridae -Pisaurinae). Rev. ZooL Afri- caines, 92:37-76. Blandin, P. 1978b. Etudes sur les Pisauridae Afri- caines. X, Description d'un Maypacius nouveau du Senegal et remarques biogeographiques sur le genre Maypacius (Araneae -Pisauridae -Pisauri- nae). Rev. ZooL Africaines, 92:658-662. Blandin, P. 1979a. Etudes sur les Pisauridae Afri- caines. XL Genres peu connus ou nouveaux des lies Canaries, du Continent Africain et de Mad- agascar (Araneae, Pisauridae). Rev. ZooL Afri- caines, 93:347-375. Blandin, P. 1979b. Cycle biologique et production de Faraignee Afropisaura valida (Simon, 1885) (Araneae -Pisauridae) dans une savane d' Afrique occidentale (Lamto, Cote dTvoire). Trop. EcoL, 20:78-93. Blandin, P. & M.-L. Celerier. 1981. Les araignees des savanes de Lamto (Cote dTvoire). Ecole Normale Superieure, Publications du Laboratoire de Zoologie 21, 2 vols., 587 pp. Bosenberg, W, & H. Lenz, 1894. Ostafrikanische Spirmen gesammelt von Herm Dr. F. Stuhlmann in den Jahren 1888 und 1889. Jahrb. hambur- gischen wiss. Anstalten, 12:27-51. Bosenberg, W. & E. Strand. 1906. Japanische Spin- nen. Abh. Senckenb. naturf. Ges., 30:93-422. Bonnet, P. 1955-1959. Bibliographia Araneoram, 2:1-5058. Brady, A. 1964. The lynx spiders of North Amer- ica, north of Mexico (Araneae: Oxyopidae). Bull. Mus. Comp. ZooL, 131:429-518. Bremer, K. 1992. Ancestral aereas: a cladistic re- interpretation of the center of origin concept. Syst. Biol., 41:436-445. Brignoli, P.M. 1983. A Catalogue of the Araneae described between 1940 and 1981 (P. Merrett, ed.). Manchester Univ. Press, Manchester. 755p. Bristowe, W.S. 1958. The world of spiders. Col- lins, London. 304 pp. Capocasale, R.M. 1982. Las especies del genero Porrimosa Roewer (Araneae, Hippasinae). J. Ar- achnoL, 10:145-156. Caporiacco, L. di, 1939. Arachnida. Pp. 303-385. In Missione biologica nel paese dei Borana. Ra- colti zoologiche, Roma. Caporiacco, L. di, 1941. Arachnida. Miss. Biol. Sagan-Omo 12 ZooL, 6:1-159. Caporiacco, L. di. 1947. Arachnida Africae orient a Dom Kittenberger, Kovacs et Bomemisza lecta. Ann. Nat. Mus. National Hungarici, 40:97-257. Carico, J.E. 1972. The Nearctic spider genus PP saurina (Pisauridae). Psyche, 79:295-310. Carico, J.E. 1973. The nearctic species of the ge- nus Dolomedes (Araneae: Pisauridae). Bull. Mus. Comp. ZooL, 144:435-488. Carico, J.E. 1976. The spider genus Tinus (Pisaur- idae). Psyche, 83:63-78. Carico, J.E. 1981. The Neotropical spider genera Architis and Staberius (Pisauridae). Bull. Amer- ican Mus. Nat. Hist., 170:140-153. Carico, J.E. 1985. Description and significance of the juvenile web of Pisaurina mira (Walck.) (Araneae: Pisauridae). Bull. British Arachnol. Soc., 6:295-296. Carico, J.E. 1986. Trechaleidae: A “new” Ameri- can spider family. P. 305. In Proc. Ninth Inter- national Congress Arachnol. (W.G. Eberhard, Y.D. Lubin & B.C. Robinson, eds.). Panama 1983, Smithsonian Press. Carico, J.E. 1993. Revision of the genus Trechalea Thorell (Araneae, Trechaleidae) with a review of SffiRWALD—PHYLOGENY OF PISAURINE SPIDERS 405 the taxonomy of the Trechaleidae and Pisauridae of the western hemisphere. J. ArachnoL, 21:226- 257. Chen, X.E. & J.C. Gao. 1990. The Sichuan farm- land spiders in China. Sichuan Sci. & Tech, Publ. House, Chengdu, 226 pp. Chen, X.E. & Z.H. Zhang. 1991. Fauna of Zhe- jiang: Araneidae. Zheijang Sci. & Tech. Publ. House., 356 pp. Chikuni, Y. 1989. Pictorial Encyclopedia of spiders in Japan. Kaisei-sha Publ. Co., Tokyo, 310 pp. Chrysanthus, P. 1967. Spiders from South New Guinea. VIII. Nova Guinea (N.S., ZooL), 35: 401-426. Coddington, J.A. & H.W. Levi. 1991. Systematics and evolution of spiders (Araneae). Ann. Rev. Ecol. Syst, 22:565-592. Comstock, J.H. 1910. The palpi of male spiders. Ann. Entomol. Soc. America, 3:161-185. Coryndon, S.C. & R.J.G. Savage. 1973. The origin and affinities of African mammal faunas. Pp. 121-135. In Special Papers in Palaeontology, No. 12. (N.E Hughes, ed.). The Palaeontol. Asso. Crowell, J.C. & L.T Frakes. 1970. Phanerozoic glaciation and the causes of ice ages. American J. Sci,, 268:193-224. Dahl, F. 1908. Die Lycosiden oder Wolfspinnen Deutschlands und ihre Stellung im Haushalte der Natur. Nova Acta -Abh. Kaiserl. Leopoldinisch- Carolinischen Deutschen Akad. Naturf., 88:1- 678. Dinesen, L., T. Lehmberg, J.O. Svendsen, L.A. Hansen & J. Fjeldsa. 1994. A new genus and species of perdicine bird (Phasianidae, Perdicini) from Tanzania; a relict form with Indo-Malayan affinities. Ibis, 136:2-11. Doleschall, C.L. 1859. Tweede Bijdrage tot de kennis der Arachniden van den Indischen Archi- pel. Naturw. Tijdschft Ned.-Indie, 13:399-434. Farris, S. 1988. Hennig86, version 1.5. Computer program and documentation distributed through Arnold Kluge, Univ. Michigan, Ann Arbor, Michigan. Gerhardt, U. & A. Kastner. 1938. Araneae. In Kii- kenthal, W. & T. Krumbach. Handbuch der Zool- ogie, 3:394-656. Griswold, C.E. 1993. Investigations into the phy- logeny of the lycosoid spiders and their kin (Arachnida, Araneae, Lycosoidea). Smithsonian Contr. ZooL, 539:1-39. Hasselt, A.W.M. van. 1884. Waamemingen om- trent Anomalien van de Geslachtsdrift bij Spin- nen-Mares., Tijd. Entomol., 27:197-206. Hasselt, A.W.M. van. 1899. Een brutengenoon spinnen-verblijf. Tijd. Entomol., 42:173-179. Hu, J.L. 1984. The Chinese spiders collected from the fields and the forests. Tianjin Press of Sci. & Tech. Huber, K.C., TS. Haider, M.W. Muller, B.A. Huber, R.J. Schweyen & EG. Barth. 1993. DNA se- quence data indicates the polyphyly of the family Ctenidae (Araneae). J. ArachnoL, 21:194-201. ICZN 1985. International code of zoological no- menclature., 3rd ed. Intern. Trust for ZooL No- men., etc., London. 338 pp. Jarvi, T.H. 1905. Zur Morphologic der Vaginalor- gane einiger Lycosiden. Festschrift fur Palmen, 6:3-36. Karsch, F. 1878. Exotisch arachnologisches. Zeitsch. ges. Naturwiss. Halle, 51:323-333. Keyserling, E. 1877. Ueber amerikanischen Spin- nenarten der Unterordnung Citigradae. Verb, zool.-bot. Ges. Wien, 26:609-708. Keyserling, E. 1891. Die Spinnen Amerikas. Bras- ilianische Spinnen, 3:1-278. Koch, L. 1878. Die Arachniden Australiens, nach der Natur beschrieben und abgebildet. Ntimberg, Pp. 969-1044. Koh, J.K.H. 1989. A guide to common Singapore spiders. Singapore Sci. Cent., 160 pp. Latreille, RA. 1804. Tableau methodique des In- sectes. Dictionnaire (Nouveau) d’histoire natu- relle, 24:129-200 [arachnids: 131-136]. Lawrence, R.F. 1927. Contributions to a knowl- edge of the fauna of South-West Africa. V. Arachnida. Ann. South African Mus., 25:1-75. Lawrence, R.F. 1942. A contribution to the araneid fauna of Natal and Zululand. Ann. Natal Mus., 10:141-190. Lawrence, R.F. 1947. A collection of Arachnida made by Dr. I. Tragardh in Natal and Zululand. Gdteborgs Kongl. Vetens. Vitter.-Samhalles Handl. (B), 5:1-41. Lehtinen, P.T. 1967. Classification of the cribellate spiders and some allied families, with notes on the evolution of the suborder Araneomorpha. Ann. ZooL Fennici, 4:199-468. Lenler-Eriksen, P. 1969. The hunting web of young Pisaura mirabilis. J. ZooL, London, 157:391- 398. Lessert, R, de, 1915. Araignees du Kilimandjaro et du Merou (suite). I. Pisauridae. Rev. Suisse ZooL, 23:1-89. Lessert, R. de, 1916. Araignees du Kilimandjaro et du Merou (suite). II. Pisauridae. Rev. Suisse ZooL, 24:565—620. Lessert, R. de, 1925. Araignees du Sud de I’Afrique. II. Rev. Suisse ZooL, 32:323-365. Lessert, R. de, 1928. Araignees du Congo recueil- lies au cours de 1’ expedition organisee par r American Museum (1909-1915). Troisieme partie. Rev. Suisse ZooL, 35:303 — 352. Lessert, R. de, 1933. Araignees d’ Angola. Rev. Suisse ZooL, 40:85-159. Lessert, R. de, 1936. Araignees de I’Afrique Or- ientale portugaise. Rev. Suisse ZooL, 43:207- 306. Maglio, V.J. 1978. Patterns of faunal evolution. Pp. 406 THE JOURNAL OF ARACHNOLOGY 603-619. In Evolution of African mammals. (V.J. Maglio & H.B.S. Cooke, eds.) Harvard Univ. Press, Cambridge, Massachusetts. Martyn, T. 1793. Aranei or the natural history of spiders including the principal parts of the well known work of English spiders by Eleasar Albin and also the whole of the celebrated publication on Swedish spiders by Charles Clerck. Acad, for illust. & print, nat. hist. London. Pp. 1-70. Merrett, P. 1988. Notes on the biology of the neo- tropical pisaurid, Ancylometes bogotensis (Key- serling) (Araneae: Pisauridae). Bull. British Ar- achnol. Soc., 7:197-201. Nentwig, W. 1985. Architis nitidopilosa, a neo- tropical pisaurid with a permanent catching web (Araneae, Pisauridae). Bull. British Arachnol. Soc., 6:295-297-303. Nitzsche, R.O.M. 1988. “Brautgeschenk” und Umspinnen der Beute bei Pisaura mirabilis, Do- lomedes fimbriatus und Thaumasia uncata (Arachnida, Araneida, Pisauridae). Verb, natur- wiss. Ver. Hamburg (NF), 30:353-394. Nixon, K.C. 1992. Clados, version 1.1. Computer program and documentation distributed by the author, L.H. Bailey Hortorium, Cornell Univ., Ithaca, New York. Olson, S.L. 1973. A classification of Rallidae. Wil- son Bull., 85:381-416. Pavesi, P. 1895. Esplorazione del Giuba e dei suoi affluent! conspiuta dal Cap. Bottego. XVIII. Ar- acnidi. Ann. Mus. Civ. Stor. Nat. Genova, 38: 151-188. Pennington, T.D. 1991. The genera of Sapotaceae. Royal Botanic Gardens, Kew & The New York Botanical Garden. Whitstable, UK. 295 pp. Petrunkevitch, A. 1928. Systema Araneorum. Trans. Connecticut Acad. Arts & Sciences, 29: 1-270. Pickard-Cambridge, F.O. 1898. On the cteniform spiders of Africa, Arabia and Syria. Proc. Zool. Soc. London, 1898:13-32. Pickard-Cambridge, O. 1885. Araneida. Scientific Results of the second Yarkand Mission; based upon the collections and notes of the late Ferdi- nand Stoliczka, Ph.D., Calcutta, 1885:1-115. Platnick, N.L 1989. Advances in spider taxonomy 1981-1987. Manchester Univ. Press, 673 pp. Platnick, N.L 1993. Advances in spider taxonomy 1988-1991. New York Entomol. Soc., in associ- ation with the American Mus. Nat. Hist., 846 pp. Pocock, R.I. 1899. On scorpions, pedipalps, and spiders from tropical West Africa, represented in the collection of the British Museum. Proc. Zool. Soc. London, 1899:833-885. Pocock, R.I. 1900. The fauna of British India, in- cluding Ceylon and Burma. Arachnida. London, 279 pp. Roewer, C. Fr. 1954. Katalog der Araneae. Inst. R. Sci. Nat. Belgique. Vol. 2a, 923 pp. Brussels. Roewer, C. Fr. 1955. Lycosaeformia 1. Exploration de Parc Nat. de TUpemba, 30:1-420. Roewer, C. Fr. 1961. IV. Opilioniden und Araneen., Vol. 2. In Le Parc National du Niokolo-Koba. Mem. ITnstitut Francais d’Afrique Noire, 62:33- 81. Savage, J.M. 1973. The geographic distribution of frogs: patterns and predictions. Pp. 351-445. In Evolutionary Biology of the Anurans. (J.L. Vial, ed.). Sierwald, P. 1987. Revision der Gattung Thalassius (Arachnida, Araneae, Pisauridae). Verb, natur- wiss. Ver. Hamburg, (NF), 29:51-142. Sierwald, P. 1989a. Genital morphology of the ge- nus Perenethis (Araneae: Pisauridae). Reports from the Dept. Biology, University in Turku, 19: 97 (abstract). Sierwald, P. 1989b. Morphology and ontogeny of female copulatory organs in American Pisauridae with special reference to homologous features (Arachnida: Araneae). Smithsonian Contr. Zool., 484:1-24. Sierwald, P. 1990. Morphology and homologous features in the male palpal organ in Pisauridae and other spider families, with notes on the tax- onomy of Pisauridae (Arachnida: Araneae). Nemouria, 35:1-59. Sierwald, R 1993. Revision of the genus Parados- sensus with notes on the family Trechaleidae. Rev. Arachnologique, 10:53-74. Simon, E. 1885. Etudes arachnologiques. 18e Me- moire. 26. Materiaux pour servir a la faune des Arachnides du Senegal. Ann. Soc. Entomol. France, (6) 5:345-396. Simon, E. 1889. Etudes arachnologiques. 21e Me- moire. 31. Descriptions d’especes et de genres nouveaux de Madagascar et de Mayotte. Ann. Soc. Entomol. France, (6) 8:237-247. Simon, E. 1890. Etudes arachnologiques. 22e Me- moire. XXXIV. Etude sur les Arachnides de r Yemen. [Remarques sur la classification des ar- aignees, pp. 79-82]. Ann. Soc. Entomol. France, (6) 10:77-124. Simon, E, 1891. Etudes arachnologiques. 23e Me- moire. 37. Arachnides recueillis sur le haut Con- go par M. Ant. Greshoff. Ann. Soc. Entomol. France, 50:297-299. Simon, E. 1897. Arachnides recueilles par M.M. Maindron a Kurrachee et a Matheran (pres Bom- bay) en 1896. Bull. Mus. Hist. Nat., 1897:289- 297. Simon, E. 1898a. Histoire naturelle des Araignees. Roret, Paris, 2:193-380. Simon, E. 1898b. Descriptions d’ Arachnides nou- veaux des families des Agelenidae, Pisauridae, Lycosidae et Oxyopidae. Ann. Soc. Entomol. Belgique, 42:5-34. Simon, E. 1909. Arachnides recueillis par L. Fea SIERWALD— PHYLOGENY OF PISAURINE SPIDERS 407 sur la cote occidentale d’Afrique. 2e partie. Ann, Mus. Civ. Stor. Nat. Genova, 4 [— 44]:335-449. Smith, A.G., D.G. Smith & B.M. Funnel. 1994. Atlas of mesozoic and cenozoic coastlines. Cam- bridge Univ, Press, 99 pp. Song, D.X. 1987. Spiders from agricultural regions of China (Arachnida: Araneae). Agriculture Pub- lishing House, Beijing. Strand, E. 1906. Diagnosen nordafrikanischer, hauptsachlich von Carlo Freiherr von Erlangen gesammelter Spinnen. Zool. Anz., 30:604-637, 655-690. Strand, E. 1908. Exotisch-araneologisches. II. Spinnen aus Kamerun. Jahrbuch des Nassauisch- en Vereins fur Naturkunde, 61:223-295. Strand, E. 1911. Araneae von den Aru-und Kei- Inseln. Abh. Senckenb. naturf. Ges., 34:129- 199. Strand, E. 1913. Arachnida. 1. In Wissenschaftliche Ergebnisse der Deutschen Zentral-Afrika Expe- dition 1907-1908, 4(ZooL 2)(II): 325-474. Strand, E. 1928. Miscellanea nomenclatorica zool- ogica et palaeontologica, I-II. Archiv fur Natur- geschichte 92 A [for 1926]: 30-75 [Arachnida, Pp. 42-45]. Thorell, T. 1881. Studi sui Ragni Males! e Papuan!, Part. III. Ragni deli Austro-Malesia e del Capo York, conservati nel Museo civico di storia na- tuale di Genova. Ann. Mus. Civ. Stor. Nat. Gen- ova, 17 (VII-XXVII): 1-720. Thorell, T. 1891. Spindlar fran Nikobarema och andra delar af Sodra Asien. Bih. Kongl. Svenska Vetens.-Akad. Hand!., 24:1-149. Thorell, T. 1895. Descriptive catalogue of spiders of Burma. British Museum, London, 406 pp. Thorell, T. 1899. Araneae Camerunenses (Africae occidentalis) quas anno 1891 collegrunt Cel. Dr. Y. Sjostedt aliique. Bih. Kongl. Svenska Vetens.- Akad. Handl., 25:1-105. Todd-Davis, V. 1985. Araneomorphae (in part). Pp. 49-125. In Zool. Cat. Australia, 3. Arachnida. Walckenaer, C.A. 1847. Histoire naturelle des In- sectes. Apteres. Tome IV. Paris, 623 pp. Workman, T. & M.E. Workman. 1897. Malaysian spiders. Part 11, 12, 13. Belfast, Pp. 81-104. [non vidi] Yaginuma, T. 1967. Three new spiders (Argiope, Boethus and Cispius) from Japan. Acta Arach- noL, 20:50-64 [in Japanese with English sum- mary]. Yaginuma, T. 1986. Spiders of Japan in color (new ed.). Hoikusha Publ. Co., Osaka. Yaginuma, T. 1991. A new genus, Shinobius, of the Japanese pisaurid spider (Araneae: Pisauri- dae). Acta ArachnoL, 40:1-6. Zahl, P.A. 1971. What’s so special about spiders. Nat. Geographic, 140:190-219. Manuscript received 26 March 1996, revised 30 November 1996. 1997. The Journal of Arachnology 25:408-409 RESEARCH NOTE A NEW GENERIC SYNONYMY IN SCORPIONS: SCORPIOBUTHUS WERNER = UROPLECTES PETERS (SCORPIONES, BUTHIDAE) The genus Scorpiobuthus was briefly de- scribed by E Werner (1939) from specimens lacking locality data, and it has been more or less forgotten since that time. The genus was monotypic, containing only the species Scor- piobuthus apatris Werner 1939, described in the same paper. Interestingly, Werner (1939) did not assign Scorpiobuthus to a family, but indicated it was close to Buthoscorpio Werner 1936 (which he regarded as a member of the Scorpionidae). The type specimens have not been subsequently studied, and no new spec- imens have been reported in the last 57 years. The genus name was rediscovered by Francke (1985) and listed among valid scorpion ge- neric names in his conspectus, under Buthi- dae. However, it was not included in recent generic keys to buthids or other families (Stahnke 1972; Sissom 1990). Through the kindness and enthusiastic sup- port of Dr. Franz Krapp, the curator of the Lower Invertebrates Division of the Zoolo- gisches Forschungsinstitut und Museum Koe- nig (Bonn, Germany), we were able to ex- amine the type specimens of S. apatris which are deposited in this Museum. The type series consists of two adult female syntypes (dried and later rehydrated; partially damaged). Nos. 82 and 83. We hereby designate No. 83 as lectotype and No. 82 as paralectotype. First of all, Scorpiobuthus is indeed a bu- thid. The sternum of Scorpiobuthus was re- ported to be subpentagonal (Werner 1939), but examination of the types reveals that the lat- eral edges of the structure are moderately con- vergent anteriorly — in fact, the sternum could be regarded as subtriangular, although not ex- tremely so. Further, as in other buthids, the anterior aspect of the sternum bears a small lobe-like structure that is separated from the main portion by a distinct groove. Upon study of additional characters, it became clear to us that the specimens are referable to Uroplectes Peters 1862. They share the following diag- nostic characters with members of that genus: (1) the alpha-pattern of dorsal trichobothria of the pedipalp femur; (2) the presence of a dis- tinct subaculear tooth; (3) the absence of den- ticles on the undersurface of the cheliceral fixed finger; (4) enlarged proximal pectinal teeth in the female (found in many Uroplec- tes); (5) the dentition pattern of the pedipalp chela fingers; (6) reduction of the carapacial carinae; and (7) the presence of tibial spurs on legs III and IV. If the sternum is regarded as subtriangular, the specimens trace easily to Uroplectes in Sissom’s (1990) key to buthid genera. We therefore propose the following synonymy: Scorpiobuthus Werner 1939 = Uroplectes Peters 1862. The genus Uroplectes is widespread in southern and eastern Africa. Checking keys published for South Africa (Hewitt 1918; Lawrence 1955), East Africa (Probst 1973) and Namibia (Lamoral 1979), we discovered a close match with Uroplectes chubbi Hirst 1911. This species is unusual in that all five metasomal segments are smooth and coarsely punctate, a feature found in the two specimens of Scorpiobuthus apatris. The specimens match other details provided in a brief de- scription of U. chubbi by Hewitt (1918). Con- sequently, we propose the following species synonymy: Scorpiobuthus apatris Werner 1939 = Uroplectes chubbi Hirst 1911. On a final note, the status of U. chubbi is somewhat uncertain and needs clarification. Hewitt (1918) suspected that U. chubbi was a junior synonym of U. jutrzenkai Penther 1900. However, at the bottom of the same page he suggested that U. chubbi had affinites with U. xanthogrammus Pocock 1897, which was sug- gested to be a “variety” of U.fischeri (Karsch 1879) by Kraepelin (1913). Hewitt then stated 408 FET & SISSOM— SCORPION SYNONYMY that U. chubbi was probably a variety of U. fischeri as well, perhaps unaware that Birula (1915) had accepted U. xanthogrammus as a valid species. More recently, U. xanthogram- mus was regarded as a subspecies of U. fis- cheri by Probst (1973), and U. jutrzenkai was synonymized with U. vittatus (Thorell 1876) by Newlands (1970). The latter author ap- peared to consider U. chubbi distinct from U. vittatus. Finally, Lamoral & Reynders (1975) recognized all three taxa {U. vittatus, U. chub- bi, and U. fischeri) as distinct species. Clearly, the situation requires further study. ACKNOWLEDGMENTS We sincerely thank Franz Krapp, the cura- tor of the Lower Invertebrates Division of the Zoologisches Forschungsinstitut und Museum Koenig (Bonn, Germany) for allowing to ex- amine the types of Scorpiobuthus apatris. We also thank Matt E. Braunwalder (Zurich, Swit- zerland) for his valuable help with biblio- graphic sources. Kari J. McWest read the manuscript, and made several valuable sug- gestions. LITERATURE CITED Birula, A.A, 1915. A general list of the scorpions of British East Africa. Scientific Results of the Zoological Expedition to British East Africa and Uganda made by Prof. V. Dogiel and I. Sokolow (St. Petersbourg), 1:1-31 (in Russian and En- glish). Francke, O.F. 1985. Conspectus genericus scor- pionorum 1758-1982 (Arachnida: Scorpiones). Occas. Pap. Texas Tech Univ., 98: 1-32. Hewitt, J. 1918. A survey of the scorpion fauna of South Africa. Trans. R. Soc. South Africa (Cape Town), 6:89-192. BQ-aepelin, K. 1913. Neue Beitrage zur Systematik der Gliederspinnen. III. A. Bemerkungen zur Skorpionenfauna Indiens. B. Die Skorpione, Pe- dipalpen und Solifugen Deutsch-Ost-Afrikas. 409 Mitteilungen aus dem Naturhistorischen Muse- um, Hamburg (2. Beiheft zum Jahrbuch der Hamburgischen Wissenschaftlichen Anstalten, 1912), 30:123-196. Lamoral, B.H. 1979. The scorpions of Namibia (Arachnida: Scorpionida). Ann. Natal Mus., (Pie- termaritzburg), 23:497-784. Lamoral, B.H. & S.C. Reynders. 1975. A catalogue of the scorpions described from the Ethiopian faunal region up to December 1973. Ann. Natal Mus., (Pietermaritzburg), 22:489-576. Lawrence, R.E 1955. Solifugae, Scorpions and Pedipalpi, with checklist and keys to South Af- rican families, genera and species. Results of the Lund University Expedition in 1950-1951. Pp. 152-259 In, South African Animal Life, Uppsala 1:152-262. Newlands, G. 1970. A re-examination of some South African scorpion species. Ann. Transvaal Mus., 26:199-210. Probst, PJ. 1973. A review of the scorpions of East Africa with special regard to Kenya and Tanza- nia. Acta Trop., 30:312-335. Sissom, W.D. 1990. Systematics, biogeography and paleontology. Chapter 3. Pp. 64-160 In, Bi- ology of Scorpions. (G.A. Polis, ed.). Stanford Univ. Press, Stanford, California. Stahnke, H.L. 1972. A key to the genera of Buthi- dae (Scorpionida). Entomol. News, 83:121-133. Werner, F. 1939. Ueber einige Scorpione aus dem Museum Alexander Koenig. Festschrift zum 60. Geburstage von Professor Dr. Embrik Strand (Riga), 5:361-362. Victor Fet: Dept, of Biological Sciences, Marshall University, Huntington, West Vir- ginia 25755 USA W. David Sissom: Dept, of Life, Earth, and Environmental Sciences, West Texas A & M University, Canyon, Texas 79016 USA Manuscript received I October 1996, accepted 15 February 1997 1997. The Journal of Arachnology 25:410-411 RESEARCH NOTE SYNONYMY OF THE PSEUDOSCORPION CHERNES INSUETUS WITH AMERICHERNES OBLONGUS (CHELONETHI, CHERNETIDAE): AN UNESTABLISHED INTRODUCTION TO BRITAIN Chernes insuetus was described by O.R- Cambridge (1892) from two specimens found in an oil mill in Dover, Kent (England). The mill was later demolished (Kew 1911) and this species has not been recorded since. Kew, who examined one of the types, noted that this species belonged to “a group with polished integuments, almost simple bristles, non-gran- ulate tergites, and with a tactile hair near ex- tremity of tibia IV.” Although this would have placed C. insuetus in the Lamprochemetinae (as then dehned), Beier (1932) listed it as a doubtful species of Allochernes Beier 1932, in which he was followed by Roewer (1937). The name insuetus did not appear again in the literature until Legg & Jones (1988) syn- onymized it with Lamprochernes chyzeri (To- mosvary 1882). Although no justihcation was given for this synonymy, it was accepted as the status quo by Harvey (1991). However, the identihcation of insuetus with chyzeri is hard to accept in view of the fact that Kew - a competent specialist - had examined British material of both species and found them to be quite distinct. In 1980 I was able to study the two female syntypes of Chernes insuetus, deposited in the Hope Entomological Collections of Oxford University Museum (HECO). The specimens were lent to a third party at the Natural His- tory Museum, London, under whose supervi- sion they were studied. Afterwards, the types were left on a desk, to be mailed the next day. Unfortunately, they disappeared before this could be done and must be presumed lost. The material of C. insuetus which Cambridge sent to E. Simon (who hrst identihed it as new to science) was evidently returned, there being no trace of this species in the collections of the Museum national d’Histoire naturelle, Paris. Although the spermathecae could not be ex- amined, the external morphology of the types of C. insuetus was found to agree with Much- more’s (1976) redescription of Americhernes oblongus (Say 1821). Chernes insuetus Cam- bridge is therefore considered to be a junior subjective synonym of A. oblongus. Americhernes oblongus (Say 1821) Che lifer oblongus Say 1821:64. Neotype d from Havana, Illinois, USA; designated by Hoff (1949) (Illinois Natural History Survey, not examined). Americhernes oblongus (Say): Muchmore 1976: 153-156, figs. 3-9; Harvey 1991:542 (complete synonymy up to 1989). Chelifer communis var. pennsylvanicus Ellingsen 1910:366 (synonymized by Muchmore 1991:80). Chelifer n. sp. Cambridge 1884:103. Chernes insuetus Cambridge 1892:225-226, pL C fig. 17; Kew 1916:130-131. Syntypes 2$, from debris and refuse in oil mill, Dover, Kent, En- gland, leg. W.P. Haydon, 1880 (HECO, examined; now lost). NEW SYNONYMY. Chelifer (Chernes) insuetus (Cambridge): Kew 1911:41 (footnote 1). AllochemesO insuetus (Cambridge): Beier 1932: 154; Roewer 1940:298. Lamprochernes chyzeri (not Tomosvary): Legg & Jones 1988:102 (in part); Harvey 1991:588 (in part). As the derivation of its junior synonym im- plies (Latin insuetus, unaccustomed), Cam- bridge (1884) regarded this species as alien to the British fauna, perhaps having been im- ported with oilseeds used in the mill. Ameri- chernes oblongus is widely distributed in the United States (Muchmore 1976; Harvey 410 JUDSON— SYNONYMY OF THE PSEUDOSCORPION 411 1991), and it is likely that the Kentish popu- lation originated from the eastern seaboard of North America. It is worth noting that they were found in company with the first known British specimens of Withius piger (Simon 1878) (syn. Chelifer subruber Simon 1879), another introduced species (Cambridge 1884, 1892). Americhernes Muchmore 1976 is currently known from the Americas, Australia and the Pacific (Muchmore 1976; Harvey 1990), but there have been no subsequent records of this genus from Europe. Although several Euro- pean pseudoscorpions have been found in North America (Muchmore 1972), this ap- pears to be the first record of an introduction in the opposite direction. In this case, how- ever, it is clear that A. oblongus did not be- come established in Britain. ACKNOWLEDGMENTS I am grateful to A. Smith for arranging the loan of the syntypes of Chernes insuetus and to I. Lansbury for his help during a recent visit to Oxford University Museum. Mark Harvey and Bill Muchmore are thanked for very help- ful reviews of the manuscript. LITERATURE CITED Beier, M. 1932. Pseudoscorpionidea II. Subord. C. Cheliferinea. Tierreich, 58:1-294. Cambridge, O.P. 1884. Pseudoscorpions new to Britain. Naturalist, London, 10:103. Cambridge, O.P. 1892. On the British species of false- scorpions. Proc. Dorset Nat. Hist, and An- tiq. Fid. Club Archaeol. Soc., 13:199-231, pis. A-C. Harvey, M.S. 1990. New pseudoscorpions of the genera Americhernes Muchmore and Cordyloch- ernes Beier from Australia (Pseudoscorpionida: Chemetidae). Mem. Mus. Victoria, 50:325-336. Harvey, M.S. 1991. Catalogue of the Pseudoscor- pionida. Manchester Univ. Press, Manchester. Hoff, C.C. 1949. The pseudoscorpions of Illinois. Bull. Illinois Nat. Hist. Surv., 24:407-498. Kew, H.W. 1911. A synopsis of the false-scorpions of Britain and Ireland. Proc. R. Irish Acad., (B) 29:38-64, pis. 4-6. Kew, H.W. 1916. An historical account of the pseudoscorpion-fauna of the British Islands. J. Quekett Micros. Club, (2) 13:117-136. Legg, G. & Jones, R.E. 1988. Pseudoscorpions (Arthropoda, Arachnida): Keys and notes for the identification of the species. Synopses British Fauna, (new ser.) 40:1-159. E.J. Brill & Dr. W. Backhuys, Leiden. Muchmore, W.B. 1972. European pseudoscorpions from New England. J. New York Entomol. Soc., 80:109-110. Muchmore, W.B. 1976. Pseudoscorpions from Florida and the Caribbean area. 5. Americhernes, a new genus based upon Chelifer oblongus Say (Chemetidae). Florida Entomol., 59:151-163. Muchmore, W.B. 1991. The identity of Chelifer communis var. pennsylvanicus and description of a new species of Lustrochernes (Pseudoscorpion- ida: Chemetidae). Entomol. News, 102:79-89. Roewer, C.E 1937. Chelonethi oder Pseudoskor- pione. In Klassen und Ordnungen des Tierreichs (H.G. Bronn, ed.) (5: Arthropoda; Section 4: Ar- achnoidea), 6(1, issue 2): 161-320. Akademische Verlagsgesellschaft M.B.H., Leipzig. Say, T. 1821. An account of the Arachnides of the United States. J. Acad. Nat. Sci. Philadelphia, 2: 59-82. Mark L.I. Judson: Museum national d’Histoire naturelle, Laboratoire de Zoolo- gie (Arthropodes), 61, rue de Buffon, 75005 Paris, France. Manuscript received 2 January 1997, accepted 15 April 1997. 1997. The Journal of Arachnology 25:412 ARACHNOLOGICAL RESEARCH FUND The AAS Fund for Arachnological Re- search (AAS Fund) is funded and adminis- tered by the American Arachnological Soci- ety. The purpose of the fund is to provide re- search support for work relating to any aspect of the behavior, ecology, physiology, evolu- tion, and systematics of any of the arachnid groups. Awards may be used for field work, museum research (including travel), expend- able supplies, identification of specimens, and/ or preparation of figures and drawings for publication. Monies from the fund are not de- signed to augment or replace salary. Individual awards will not exceed $1000.00, and, although open to all students and faculty with less than $500.00 per year research budget, preference will be given to students. A total of $6000.00 is available for awarding during each funding year. A funding year has two granting periods: Winter-Spring and Summer-Fall. Available monies could be expended for three large proposals during each granting period, or a greater number of partially funded proposals, and/or a number of smaller, less expensive proposals. The final funding pattern is at the discretion of the re- view committee. Applications for support should be received by the chair of the review committee no later than May 30 or November 30, for funding by June 30 and December 30, respectively. To be considered for an award from the AAS Fund, please submit four copies of a proposal of no more than five pages (including references) detailing your research project. Proposals should have three main parts: 1) an INTRODUCTION where background in- formation is presented relative to the proposed work. The introduction should include a sec- tion which places the proposed work in con- text with currently known relevant informa- tion, a section which provides justification for the proposed work, and a clear statement of the hypothesis(ses) to be tested, or, in the case of systematic revisions, the type of synthesis that will be achieved and its significance; 2) a METHODS section where the methods, ma- terials, experimental design, and statistical or taxonomic analysis(ses) to be used are clearly and concisely presented, and 3) a BUDGET explaining (in detail) how monies awarded will be spent in the proposed research. Proposals should be submitted to: Dr. Elizabeth M. Jakob, AAS Fund Chair Department of Entomology 102B Femald Hall University of Massachusetts Amherst, MA 01003-2410 USA Proposals must be submitted in English. The four copies of the proposal must be in the hands of the Fund chair by the appropriate deadlines to be considered. Electronic sub- mission and/or FAX submission of proposals with hard copies to follow is acceptable only if the other copies arrive before the stated deadlines. If these submission rules are diffi- cult or prohibitive because of cost, erratic postal services, or remote location (remote field stations or sites), other methods of sub- mission may be acceptable. For other submis- sion possibilities, please contact the chair of the Fund at the above address, or electroni- cally at ejakob@bgnet.bgsu.edu. Alternative submissions will be accepted only if the chair has been previously contacted, and all dead- lines will still apply. 412 INSTRUCTIONS TO AUTHORS (revised October 1996) Manuscripts are preferred in English but may be ac- cepted in Spanish, French or Portuguese subject to availability of appropriate reviewers. Authors whose pri- mary language is not English may consult the Associate Editor for assistance in obtaining help with English manuscript preparation. All manuscripts should be pre- pared in general accordance with the current edition of the Council of Biological Editors Style Manual unless instructed otherwise below. Authors are advised to con- sult a recent issue of the Journal of Arachnology for additional points of style. Manuscripts longer than 1500 words should be prepared as Feature Articles, shorter papers as Research Notes. Send four identical copies of the typed material together with photocopies of illustra- tions to the Associate Editor. Do not send original handmade illustrations until the manuscript has been ac- cepted. Mail to: Petra Sierwald, Associate Editor; Division of In- sects, Dept, of Zoology, Field Museum, Roosevelt Road at Lakeshore Drive, Chicago, IL 60605 USA [Telephone: (312) 922-9410, ext. 841; FAX: (312) 663-5397; E-mail: SIERWALD@FMNH.ORG] Correspondence relating to the initial submission of a manuscript, as well as the review process, should be directed to the Associate Editor. Correspondence re- lating all other matters should be directed to the Editor. After the manuscript has been accepted, the author will be asked to submit the manuscript on a computer disc. It must be in a widely used program (preferably in MS DOS WordPerfect). Indicate clearly on the computer disc the word processing program used. FEATURE ARTICLES Title page. — The title page will include the com- plete name, address, and telephone number of the au- thor with whom proofs and correspondence should be exchanged, a FAX number and electronic mail address if available, the title in capital letters, and each author’s name and address, and the running head (see below). Abstract. — The heading in capital letters should be placed at the beginning of the first paragraph set off by a period. In articles written in English, a second abstract in an acceptable language may be included pertinent to the nationality of the author(s) or geo- graphic region(s) emphasized in the article. A second abstract in English must be included in articles not written in the latter language. Text. — Double-space text, tables, legends, etc. throughout. Three levels of heads are used. The first level (METHODS, RESULTS, etc.) is typed in capitals and on a separate line. The second level head begins a paragraph with an indent and is separated from the text by a period and a dash. The third level may or may not begin a paragraph but is italicized and sepa- rated from the text by a colon. Use only the metric system unless quoting text or referencing collection data. All decimal fractions are indicated by the period regardless of language of the text. Citation of references in the text: Cite only papers already published or in press. Include within parenthe- ses the surname of the author followed by the date of publication. A comma separates multiple citations by the same author(s) and a semicolon separates citations by different authors, e.g., (Smith 1970), (Jones 1988; Smith 1993), (Smith 1986, 1987; Smith & Jones 1989; Jones et al. 1990). Literature cited section. — Use the following style: Lombardi, S. J. & D. L. Kaplan. 1990. The amino acid composition of major ampullate gland silk (dragline) of Nephila clavipes (Araneae, Tetragnath- idae). J. Arachnol., 18:297-306. Krafft, B. 1982. The significance and complexity of communication in spiders. Pp. 15-66, In Spider Communications: Mechanisms and Ecological Sig- nificance. (P. N. Witt & J. S. Rovner, eds.). Princeton University Press, Princeton, New Jersey. Footnotes. — Footnotes are permitted only on the first printed page to indicate current address or other information concerning the author. All footnotes are placed together on a separate manuscript page. Running head. — The author sumame(s) and an ab- breviated title should be typed all in capital letters and must not exceed 60 characters and spaces. The running head should be placed near the top of the title page. Taxonomic articles. — Consult a recent taxonomic article in the Journal of Arachnology for style or con- tact the Editor. Tables. — Each table, with the legend above, should be placed on a separate manuscript page. Only hori- zontal lines (usually three) should be included. Use no footnotes; instead, include all information in the leg- end. Make notations in the text margins to indicate the preferred location of tables in the printed text. Illustrations. — Address all questions concerning il- lustrations to: James W. Berry, Editor; Dept, of Biological Sci- ences, Butler University, Indianapolis, Indiana 46208 USA [Telephone (317) 940-9344; FAX (317) 940-9519; E-mail: BERRY@BUTLER.EDU] All art work must be camera-ready for reproduction. In line drawings, pay particular attention to width of lines and size of lettering when reductions are to be made by the printer. Multiple photos assembled on a single plate should be mounted with only a minimum of space separating them. In the case of multiple illus- trations mounted together, each illustration must be numbered sequentially rather than given an alphabetic sequence. Written on the back should be the name(s) of author(s) and an indication of top edge. The author should indicate whether the illustrations should be one column or two columns in width. The overall dimen- sions of original artwork should be no more than 1 1 inches (28 cm) X 14 inches (36 cm). Photocopies in review manuscripts should be reduced to the exact measurements that the author wants to appear in the final publication. Larger drawings present greater dif- ficulty in shipping and greater risks of damage for which the JOA assumes no responsibility. Make no- tations in the text margins to indicate the preferred position of illustrations in the printed text. Legends for illustrations should be placed together on the same page(s) and separate from the illustrations. Each plate must have only one legend, as indicated below: Figures 1-4. — A-us x-us, male from Timbuktu. 1, Left leg; 2, Right chelicera; 3, Dorsal aspect of geni- talia; 4, Ventral aspect of abdomen. Figures 27-34. — Right chelicerae of species of A-us from Timbuktu. 27, 29, 31, 33, Dorsal views; 28, 30, 32, 34, Prolateral views of moveable finger; 27, 28, A- us x-us, holotype male; 33, 34, A-us y-us, male. Scale = 1.0 mm. Assemble manuscript for mailing. — Assemble the separate sections or pages in the following sequence: title page, abstract, text, figure legends, footnotes, ta- bles with legends, figures. Page charges, proofs and reprints. — There are no page charges, but authors will be charged for changes made in the proof pages. Authors will receive a reprint order form along with their page proofs. Reprints will be billed at the printer’s current schedule of costs. RESEARCH NOTES Instructions above pertaining to feature articles ap- ply also to research notes, except that abstracts and most headings are not used and the author’s name and address follow the Literature Cited section. CONTENTS The Journal of Arachnology Volume 25 Feature Articles Number 3 Re-description of Togwoteeus biceps (Arachnida, Opiliones, Sclerosomatidae) with Notes on its Morphology, Karyology and Phenology by Robert G. Holmberg and James C. Cokendolpher 229 Notes on the Taxonomy of Some Old World Scorpions (Scorpiones: Buthidae, Chactidae, Ischnuridae, Scorpionidae) by Victor Fet 245 Description of the Male of Diplocentrus lourencoi (Scorpiones, Diplocentridae) by Kari J. McWest 251 Guerrobunus vallensis, a New Species of Harvestman (Opiliones: » Phalangodidae), from a Cave in Valle de Bravo, State of Mexico, Mexico by Ignacio M. Vazquez and James C. Cokendolpher 257 Kaira Is a Likely Sister Group to Metepeira, and Zygiella Is an Araneid (Araneae, Araneidae): Evidence from Mitochondrial DN A by William H. Piel and Karen J. Nutt 262 1 Scharffia, A Remarkable New Genus of Spiders from East Africa (Araneae, Cyatholipidae) by Charles E. Griswold 269 Growth Rates in the Scorpion Pseudouroctonus reddelli (Scorpionida, Vaejovidae) by Christopher A. Brown 288 A Comparison of Capture Thread and Architectural Features of Deinopoid and Araneoid Orb- Webs by Brent D. Opell 295 Natural History and Copulatory Behavior of the Spiny Orbweaving Spider Micrathena gracilis (Araneae, Araneidae) by Todd C. Bukowski and Terry E. Christenson 307 Mating Behavior of Physolimnesia australis (Acari, Limnesiidae), a Non- parasitic. Rotifer-eating Water Mite from Australia by Heather C. Proctor 321 On the Abundance and Phenology of Palpigradi (Arachnida) from Central Amazonian Upland Forests by J. Adis, U. Scheller, J.W. de Morais, B. Conde and J.M.G. Rodrigues 326 Ground-layer Spiders (Araneae) of a Georgia Piedmont Floodplain Agroecosystem: Species List, Phenology and Habitat Selection by Michael L. Draney 333 Effects of Prey Supplementation on Survival and Web Site Tenacity of Argiope trifasciata (Araneae, Araneidae): a Field Experiment by Bonnie Jean McNett and Ann L. Rypstra 352 Phylogenetic Analysis of Pisaurine Nursery Web Spiders, with Revisions of Tetragonophthalma and Perenethis (Araneae, Lycosoidea, Pisauridae) by Petra Sierwald 361 Research Notes A New Generic Synonymy in Scorpions: Scorpiobuthus Werner = Uroplectes Peters (Scorpiones, Buthidae) by Victor Fet and W. David Sissom 408 Synonymy of the Pseudoscorpion Cherries insuetus with Americhemes oblongus (Chelonethi, Chemetidae): an unestablished introduction to Britain by Mark L. I. Judson 410 Announcement Arachnological Research Fund 412 ) i: I ) I.. *) 1 ) I