AMERICAN MUSEUM Novitates PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y. 10024 Number2897, pp. 1-11, figs. 1-13 November 9, 1987

Archaeodictyna ulova, new species (Araneae: ), A Remarkable Kleptoparasite of Group-Living Eresid ( spp., Araneae: Eresidae)

CHARLES E. GRISWOLD' AND TERESA MEIKLE-GRISWOLD2

ABSTRACT A new species of Dictynidae, Archaeodictyna as an inquiline in the nest interior, and feeds com- ulova, was found living in the communal retreats munally with the eresids on prey items which the of two species of group-living eresid spiders, eresids catch. A. ulova was never found living in- Pavesi and Stegodyphus dependently or with spiders other than the two dumicola Pocock, in Natal and the eastern Trans- eresids, and did not remain in nests which were vaal regions ofSouth . A species description abandoned by the eresid hosts. Ratios ofA. ulova and diagnosis are given, with observations of be- to eresids in nests were variable, but reached as havior in the field and laboratory. The dictynid high as 0.22. Scelionid wasps parasitized eggs of lays its egg sacs throughout the eresid retreats, lives A. ulova within the eresid nests. INTRODUCTION Kleptoparasitic (prey-stealing) interactions dae (Vollrath, 1979), Theridiidae (Kaston, among spiders are widely known, with klep- 1965; Vollrath, 1984; Whitehouse, 1986), and toparasites recognized from at least five fam- Uloboridae (Bradoo, 1979). In all previously ilies: Mysmenidae (Griswold, 1985; Platnick known cases, the kleptoparasites rely on and Shadab, 1978), Salticidae (Jackson, 1985; stealth to avoid detection and attack by their Jackson and Blest, 1982), Symphytognathi- hosts, or are so small relative to the host as

' Kalbfleisch Research Fellow, Department of Entomology, American Museum of Natural History. 2 % Department of Entomology, American Museum of Natural History.

Copyright © American Museum of Natural History 1987 ISSN 0003-0082 / Price $1.60 2 AMERICAN MUSEUM NOVITATES NO. 2897 to apparently escape detection (Vollrath, A. ulova is described in this paper. This is 1984, 1987). Ofgreat interest then is the dis- the first record of kleptoparasitism in the covery of a kleptoparasitic which in- family Dictynidae, and the first record of a filtrates and integrates into colonies of com- spider kleptoparasite that lives as an inqui- munal nonterritorial spiders, and which is line and is apparently accepted as a member apparently treated as a conspecific by its hosts of a host colony of group-living spiders. (Meikle-Griswold, 1986). In conjunction with the study ofthe natural history oftwo species of group-living eresid spiders (Stegodyphus MATERIALS AND METHODS mimosarum Pavesi and S. dumicola Po- Nest associates ofthe eresid spiders Stego- cock), a new species of Archaeodictyna Ca- dyphus mimosarum and Stegodyphus dumi- poriacco (Dictynidae) was found occupying cola were studied as part ofa larger study on the Stegodyphus retreats and sharing the prey. the phenology and colony survival of these S. mimosarum and S. dumicola (also two eresid species (male and female voucher known as Magunia dumicola, sensu Lehti- specimens of each species are deposited in nen, 1967) are both group-living (Dewar and the American Museum of Natural History, Koopowitz, 1970; Wickler, 1973). They are New York; British Museum (Natural Histo- communal nonterritorial (sensu Jackson, ry), London; California Academy ofSciences, 1978), and cooperative (sensu Smith, 1982). San Francisco; and Musee Royal de L'Afri- Both species are obligately group-living as que Centrale, Tervuren). Field observations juveniles, and facultatively group-living and of A. ulova were made monthly from their quasisocial as adults (sensu Wilson, 1971; first discovery in October 1985, until October Meikle-Griswold and Griswold, in prep.). 1986, at the Spioenkop Dam Nature Reserve, Their distributions overlap throughout sum- in eastern South Africa (28041'S: 29°28'E, mer rainfall areas of , but S. elev. 900 m). Feeding behavior and interac- mimosarum is more often found in proximity tions with Stegodyphus hosts were observed to water. Colonies (groups of individuals in situ. Habitats at Spioenkop where A. ulova sharing a retreat) are usually found in thorny was abundant were searched monthly for evi- shrubs or trees, where they construct dense dence of this species living independently or silken communal retreats (nests), surrounded as a kleptoparasite with other spiders. by cribellate trap-webbing. Nest interiors are and S. mimosarum composed ofinterconnected silken tubes. The nests were collected by the authors from nests are variably shaped, usually ovoid, with Spioenkop, Mkuzi Game Reserve (northern the longest dimension ranging from 3 to 50 Natal, 27°40'S: 3201 1'E, elev. 130 m), Kruger cm. Nests are often found unoccupied by host Park (eastern Transvaal, 25°07'S: 31031'E, spiders. elev. 550 m), the Cederberg (32°32'S: 1901 7'E, Nests of S. dumicola and S. mimosarum elev. 940 m), and Kamiesberg (30°18'S: were often found to have kleptoparasitic spi- 18005'E, elev. 1220 m) mountains (western ders such as Argyrodes sp. (Theridiidae) and Cape Province); large numbers of nests were Portia sp. (Salticidae) in and around the trap- also collected at Etosha Park in Namibia (1 8- web. These caught small prey in the eresid 19°S: 15-17°E, elev. 1000-1200 m) by E. web, retreated to the edge ofthe web to feed, Griffin ofthe Windhoek State Museum. Nests and never entered the host retreats (i.e., pil- were dissected, and counts were made of the ferers, sensu Vollrath, 1984). Archaeodictyna Stegodyphus inhabitants and all nest associ- ulova lives as an inquiline within colonies of ates. S. mimosarum and S. dumicola (Meikle- Individuals of A. ulova were reared and Griswold, 1986). A. ulova is completely tol- observed in the lab, using several techniques: erated by its hosts, and often builds up to (1) A. ulova and S. mimosarum or S. dumic- large numbers in host colonies. The dictynids ola were reared together in glass shell vials interact closely with the Stegodyphus hosts, or in 9 cm diam. petri dishes, with folded and apparently are not recognized as "differ- paper strips or corks included as a substrate ent" when they join in group-feeding on prey for silk attachment. The spiders were fed fruit items (i.e., peculators, sensu Vollrath, 1984). flies (Drosophila melanogaster), mealworms 1 987 GRISWOLD AND MEIKLE-GRISWOLD: ARCHAEODICTYNA ULO VA 3

(Tenebrio molitor larvae), cockroaches (Bla- TABLE 1 tella germanica), or gryllid crickets. (2) A. Host-Occupied Stegodyphus Nests Containing A. ulova was reared in webs constructed in vials ulova Inquilines, Collected at Spioenkop and Dis- or petri dishes by S. mimosarum or S. du- sected micola, but with the eresids subsequently re- moved. These spiders were fed live Drosoph- Collection No. spiders in nest ila. (3) A. ulova was reared in groups of Host species date Host A. ulova conspecifics only, in glass vials containing pa- S. mimosarum Jan. 1986 294 64 per strips for silk attachment. Here they were S. mimosarum Feb. 1986 156 24 fed incapacitated fruit flies, as the web in these S. mimosarum Mar. 1986 161 1 vials was not sufficient to entangle active Dro- S. mimosarum Apr. 1986 75 1 S. mimosarum May 1986 100 26 sophila. All of the rearing setups were given S. mimosarum Oct. 1986 119 18 drops of water every 2 to 3 days. S. dumicola Oct. 1985 121 9 In the descriptions all abbreviations are S. dumicola Nov. 1985 26 2 standard for the Araneae. All measurements are in millimeters.

OBSERVATIONS laid on and in the eresid nests were parasit- A. ulova was found in nests of S. mimo- ized by small scelionid wasps (Idris sp., det. sarum and S. dumicola at the Spioenkop Dam L. Masner). We should note that Stegodyphus Nature Reserve, and in a nest of S. mimo- egg sacs (N > 100) were never found to con- sarum from the Kruger National Park. No tain scelionid egg parasitoids. specimens were found in S. dumicola nests A. ulova and its eresid hosts spend most of collected in the western Cape Province, or in the daylight hours within the silken tubes of Etosha Park, Namibia, or in one S. mimo- the nest core (figs. 1, 2). Just after dusk, when sarum nest collected from Mkuzi Game Re- the host eresids emerge from the nests to ex- serve. tend and repair the trap-web, A. ulova spiders A. ulova occurred year round at Spioenkop, also come out onto the web. They appear to in S. dumicola and S. mimosarum nests we put down drag lines as they wander over the examined monthly for 15 months. They were trap-webs, but they were never seen to card patchily distributed, mainly in areas with the out cribellate silk. Individuals of A. ulova highest densities of S. mimosarum colonies, never made cribellate silk in mixed-species for which they showed an apparent predilec- laboratory colonies, or when maintained in- tion. Of nests that were dissected from a va- dividually or in groups of A. ulova only, al- riety of localities at Spioenkop, 54.5 percent though they do have a well developed cal- of host-occupied S. mimosarum nests (6 of amistrum and cribellum. 1 1) contained A. ulova, whereas only 6.5 per- During locomotion, A. ulova frequently cent ofhost-occupied S. dumicola nests (2 of waved the first pair oflegs and tapped on the 31) did (table 1). A. ulova was never found web. Physical contacts with other spiders, living independently or with other spiders, both hosts and conspecifics, usually resulted and did not remain in nests which contained in briefly increased tapping on the body or no living eresid hosts. Ofthe nests dissected, legs of the spider contacted. These tapping 42.9 percent of host-unoccupied S. mimo- movements are similar to those described by sarum nests (3 of 7), and 4.3 percent ofhost- Jackson (1979) for the dictynids Mallos gre- unoccupied S. dumicola nests (3 of69) showed galis, M. trivittatus, and calcarata. previous A. ulova occupation, as evidenced When an A. ulova individual approached a by old A. ulova egg sacs. Such egg sacs were potential feeding situation, other spiders, or laid throughout the host nests, both inside potential prey items, its tapping pattern be- the interior tunnels of the nest core (fig. 3), came more rapid and jerky. and on the outside of the nest core. Egg sacs Feeding behavior was observed both in the contained from 1 to 10 eggs (x = 5.7, N = field and in lab colonies. In the field, prey 96 egg sacs). Many of the A. ulova egg sacs entangled in the Stegodyphus trap-web at- 4 AMERICAN MUSEUM NOVITATES NO. 2897

- I a.. A I

"I. v, 1;.

Figs. 1-4. Archaeodictyna ulova, n. sp., in nests ofStegodyphus mimosarum Pavesi from Spioenkop, Natal, South Africa. 1. Adult S. mimosarum, juvenile S. mimosarum, and adult A. ulova (arrows) at entrance to tube on surface of nest. 2. A. ulova adults at nest entrance tube. 3. Interior of nest with egg- sacs of A. ulova (arrow) and S. mimosarum, with juvenile of S. mimosarum. 4. Mass feeding on sarcophagid fly on nest surface by S. mimosarum and A. ulova (arrows). tracted both A. ulova and Stegodyphus from prey item before a Stegodyphus encountered the interior ofthe nest (fig. 4). In 9 of 10 cases it. The A. ulova in this case did not attempt recorded, Stegodyphus was the first to attack to bite the prey. On average, A. ulova waited and subdue the prey, while A. ulova remained 18 minutes to begin to feed after S. mimo- several centimeters away from the attack. Af- sarum had begun (N = 7, range 5-24 min). ter the prey was subdued and Stegodyphus A. ulova frequently continued to feed after its was feeding, A. ulova walked over the feeding hosts had stopped. Stegodyphus until an empty place was found In the lab, A. ulova living with colonies of where they could feed or until a Stegodyphus Stegodyphus mimosarum or S. dumicola moved and an A. ulova could take its place showed different behaviors depending on the (fig. 4). In one case, an A. ulova touched the size of the prey. Large prey such as cock- 1987 GRISWOLD AND MEIKLE-GRISWOLD: ARCHAEODICTYNA ULOVA 5 roaches or crickets was often noticed and colonies. Dispersal and host-finding behav- touched first by A. ulova, but was never at- iors require more study. tacked or subdued by the kleptoparasites. The prey sometimes began to struggle more vig- orously in the web when A. ulova touched Archaeodictyna ulova, new species and this often attracted them, Stegodyphus Figures 5-13 which would then attack the prey. A. ulova would begin to feed after Stegodyphus had TYPES: Holotype male, South Africa, Na- subdued the prey. In lab colonies, it was ap- tal, Spioenkop Dam, south shore, 30 km SW parent that A. ulova, by walking on the hosts Ladysmith, elev. 900 m (284 1'S: 29°28'E), and tapping them rapidly with the forelegs, from nest of Stegodyphus mimosarum in often caused Stegodyphus to shift feeding po- Acacia karroo in mixed grasslands and dry sitions or to leave the prey. bushveld, 9 July 1986, T. Meikle-Griswold When individuals of Drosophila melano- and C. Griswold, deposited in the Natal Mu- gaster were put into small colonies of Stego- seum (type no. 3391). Paratypes, same lo- dyphus and A. ulova in the laboratory, A. cality, various dates, from nests of Stegody- ulova often approached, touched, and bit the phus mimosarum and S. dumicola, 18 males, Drosophila before the Stegodyphus did. Ifthe 12 females: 1 pair each in British Museum Drosophila did not struggle after the initial (Natural History), London; National Collec- bite, Stegodyphus often did not notice the tion, Pretoria; National Museum of Zim- potential prey. A. ulova then fed either in- babwe, Bulawayo; Musee Royal de L'Afrique dividually or in groups of up to six on the Centrale, Tervuren; National Museum of Drosophila. When a single A. ulova made the Natural History, Washington, D.C.; Califor- initial bite, other A. ulova often joined within nia Academy ofSciences, San Francisco; and 1-10 minutes. Pure A. ulova colonies in vials American Museum of Natural History, New with recently made Stegodyphus webbing York; remainder in Natal Museum. could attack, subdue, and feed on Drosophila ETYMOLOGY: The specific name is from the which became entangled. A. ulova noticed Zulu word for loafer or one who lives off the and encountered, but did not bite, larger prey work of others. items. Pure A. ulova colonies in vials without DIAGNOSIS: Archaeodictyna ulova may be Stegodyphus webbing made loose, smooth distinguished from all other African species sheets from their draglines. Drosophila flies currently (Lehtinen, 1967) placed in Archaeo- rarely became entangled in this webbing, and dictyna by details of the male palpus (figs. 8, nearly always escaped before any A. ulova 9, 12, 13) and internal genitalia ofthe female individuals were able to attack them. Dam- (fig. 7). Differentia of the other African Ar- aged Drosophila individuals which were un- chaeodictyna are as follows: anguiniceps able to walk were attacked and fed upon by (Simon) 1899-male carapace elongate an- the A. ulova. teriorly, female genitalia with membranous Egg sacs of A. ulova required from 16 to bursae narrower, sclerotized receptacula rel- 21 days (N = 5) to hatch in the lab (at ca. atively larger; condocta (O. P. Cambridge) 70-77°F). Spiderlings matured in about 5 to 1876-apex ofconductor broader and lateral 7 weeks. There was no parental care of egg chamber of receptaculum very large; conse- sacs or offspring. Females laid egg sacs at 12- cuta (O. P. Cambridge) 1872-palpal conduc- 20 day intervals, usually laying a total of 2 tor broader for entire length, tibial ctenidium or 3. Newly hatched A. ulova spiderlings fed longer, and bursae broader; gertschi (Berland on prey, but were too small to subdue even and Millot) 1939-embolic base subapical; an incapacitated Drosophila. Generations longipes (Berland) 1914-apex of conductor overlapped, although in lab colonies, parents enlarged, strongly curved, extending to base generally died at about the time their first of tibia, female with dorsolateral dark bands offspring became sexually mature. on abdomen; montana (Tullgren) 1910- em- Dispersal was never observed in nature. In bolic origin more basal and apex ofconductor the lab, A. ulova individuals of all stages broader; tazzeiti Denis 1954-ctenidia of abandoned dwindling and dying Stegodyphus palpal tibia longer, bifid; tullgreni (Caporiac- 6 AMERICAN MUSEUM NOVITATES NO. 2897

2 ~~~7

_ ,

t

A ll

Figs. 5-10. Archaeodictyna ulova, n. sp. 5. Carapace of female, dorsal. 6. Dorsum of metatarsus IV of female, showing setae and trichobothrial base. 7. Spermathecae, cleared, dorsal view. 8. Male palpus, ventral. 9. Male paipus, apex of conductor. 10. Tarsus IV of female, claw. co) 1949-palpal ctenidium longer, apex of longitudinal band and 3 pairs of faint, dark, conductor spiniform, reflexed. posterodorsal spots, venter unmarked; MALE (holotype): Total length 1.76. Car- sparsely covered with dark hairs. ,apace brown on cephalic region, thoracic fo- Carapace 0.71 long, 0.61 wide, 0.25 high vea, and striae radiating from foveal region, at OA; thoracic fovea a broad, shallow yellow-brown surrounding fovea, between depression; clypeus low, 0.04 high, height 4/5 lateral striae, and posteriorly; sparse white diameter AME. Ocular area 0.31 wide, 0.12 hairs on cephalic region and between eyes. long, width PER 1.19 times AER; ocular Chelicerae brown, unmarked; venter, legs, and quadrangle slightly narrowed in front; dis- palpi pale yellow-brown, unmarked, all tance AME-AME equal to AME diameter, sparsely covered with dark hairs. Abdomen AME-ALE 1/3 AME, PME-PME equal to yellow-white, with faint anteromedian dark PME, PME-PLE 4/5 PME, lateral eyes nearly 1 987 GRISWOLD AND MEIKLE-GRISWOLD: ARCHAEODICTYNA ULOVA 7

11

Figs. 1 1-13. Archaeodictyna ulova, n. sp., scale lines = 0.1 mm (upper scale line for fig. 1 1, lower for figs. 12, 13). 11. Female epigynum. 12. Left male palpus, ventral. 13. Left male palpus, retrolateral. contiguous; ratio of eyes AM:AL:PM:PL, Abdomen 1.12 long, 0.86 wide. Cribellum 1.13:1.25:1.13:1.00, diameter PME 0.04. and calamistrum nonfunctional, reduced. Chelicerae 0.39 long, vertical, slightly swol- Palpus with short dorsal ctenidium on tibia len at base with small lateral boss, weakly (fig. 13). Bulb as in figures 8, 12, 13; embolus carinate and bowed medially; fang furrow with slender, arising proximally on anterolateral 2 small teeth on promargin, 1 on retromargin; side of bulb, slender for entire length, apex edge of fang weakly serrate. Sternum domed, obscured by conductor; conductor thickened smooth, margin entire, truncate posteriorly, distally, apex (fig. 9) bent mesally, then ba- 0.44 long, 0.45 wide; labium 0.17 long, 0.19 sally. wide; palpal coxae converging but apices sep- Variation. Total length 1.63-1.84. Cara- arate, 0.21 long, 0.13 wide. pace markings faint, yellow-brown and clear, Leg formula 1243. Legs, palpi, and body or obscure, dark brown; abdomen with sparsely covered with plumose hairs (fig. 6). anterodorsal median dark band or pair of Spines absent. Trichobothria: 2 dorsal near small spots, abdominal markings faint to bold, base of tibiae, 2 distal on metatarsi I and II, venter unmarked or with paired dark longi- 1 distal on metatarsi III and IV; bothrium tudinal marks anterolaterad ofspinnerets and with posterior hood, trichome plumose (fig. laterad of book-lungs. Carapace length 1.16- 6). Claws pectinate (fig. 10), STC with 7-9 1.21 times width, height 0.42-0.50 width; teeth, ITC with 3-5. Measurements of legs: ocular area 2.00-2.48 times wider than long; distance AME-AME 4/5-equal to AME di- II III IV Palp ameter, AME-ALE 1/3-2/5 AME, PME-PME Femur 0.60 0.55 0.46 0.59 0.26 3/4-equal to PME, PME-PLE 3/4-equal to PLE. Patella 0.24 0.22 0.22 0.22 0.10 Clypeal height AME diameter, chelic- Tibia 0.51 0.43 0.32 0.39 0.08 2/3_4/5 Metatarsus 0.43 0.38 0.35 0.41 erae length 8-11 times clypeal height. Ster- Tarsus 0.37 0.33 0.31 0.25 0.37 num length 0.98-1.02 times width; labium Total 2.15 1.91 1.66 1.86 0.81 width 1.12-1.36 times length. 8 AMERICAN MUSEUM NOVITATES NO. 2897

FEMALE (paratype): Total length 2.04. DISCUSSION Markings and structure essentially as in males, abdominal markings fainter. Ecological and behavioral factors favoring Carapace (fig. 5) 0.86 long, 0.73 wide, 0.33 kleptoparasitism were listed by Brockmann high at OA; clypeus 0.05 high. Ocular area and Barnard (1979) in a review of 0.34 wide, 0.14 long; distance AME-AME kleptoparasitism in birds. These included: (1) equal to AME diameter, AME-ALE 2/3 AME, high host concentrations, (2) large quantities PME-PME equal to PME, PME-PLE 1¼/4 of food available, (3) large or high-quality times PME; ratio of eyes AM:AL:PM:PL, food items, (4) habits ofhost that make food 1.1 1:1.00:1.1 1: 1.22, diameter PME 0.05. temporally or spatially predictable, (5) long Chelicerae 0.39 long, vertical, noncarinate prey-handling time by host that makes prey and weakly concave medially. Sternum 0.56 more vulnerable to kleptoparasitism, (6) little long, 0.50 wide; labium 0.17 long, 0.23 wide; likelihood of host escape or retaliation, and palpal coxae 0.25 long, 0.13 wide. (7) kleptoparasites moving with agility among Leg formula 1423. Palpal claw pectinate, their hosts. These conditions all apply to the with 6 teeth. Calamistrum a single row, pos- living habits of S. mimosarum and S. du- terodorsal, extending subbasally to subapi- micola and the kleptoparasite A. ulova. (1) cally for most of metatarsus length. Mea- Host concentrations are locally high. Colo- surements of legs: nies of S. mimosarum and S. dumicola are patchily distributed, but populations are dense II within colonies and colony clusters. We found I III IV Palp S. mimosarum and S. dumicola nests to be Femur 0.76 0.69 0.59 0.71 0.29 Patella 0.29 0.27 0.25 0.29 0.12 most abundant in areas of high prey density: Tibia 0.59 0.47 0.39 0.45 0.16 e.g., near watercourses and lakes, or in areas Metatarsus 0.55 0.49 0.41 0.51 with many large grazing mammals. At the Tarsus 0.41 0.37 0.31 0.35 0.32 Spioenkop site, A. ulova was most frequently Total 2.60 2.29 1.95 2.31 0.89 found in a lakeshore area with the highest density of S. mimosarum nests. (2) These Abdomen 1.22 long, 0.96 wide. Cribellum colonies were nearly always found either han- small, entire. Epigynum (fig. 11) unsclero- dling prey or with recent prey items dead on tized externally, receptacula visible, with the nest exterior, suggesting that large quan- paired copulatory pores. Spermathecae as in tities offood were available. (3) A wide spec- figure 7, with large anterior membranous bur- trum of prey types were captured, ranging sae, small posterodorsal sclerotized receptac- from small parasitic Hymenoptera to large ula, and lateral fertilization ducts. grasshoppers and mantids. In the field, we Variation. Total length 1.80-2.90. Varia- observed A. ulova feeding alongside S. mi- tion in markings as in male. Carapace length mosarum on prey items generally at least two 1.18-1.24 times width, height 0.36-0.46 times the size of S. mimosarum. However, width; ocular area 2.29-2.54 times wider than smaller prey items were often pulled into the long; distance AME-AME 4/5-equal to AME nest interior by Stegodyphus, and in these diameter, AME-ALE 2/53/5 AME, PME-PLE cases, feeding was impossible to observe. (4) 4/5-11/3 times PME. Clypeal height about equal Food is spatially predictable from the point to AME diameter; chelicerae length 7-8 times of view ofA. ulova (i.e., on trap-web or nest clypeal height. Sternum length 0.98-1.12 surface), as long as the host spiders continue times width; labium width 1.29-1.35 times to repair and extend the trap-web and catch length. Leg formula 1423 or 1(2 = 4)3. prey. (5) Long prey-handling time makes prey ADDITIONAL MATERIAL EXAMINED: SOUTH vulnerable to kleptoparasitism. After being AFRICA: Transvaal: 30 km SW Skukuza on subdued, large prey items were usually fed Nahpe Road, Kruger National Park, elev. upon for more than one hour. (6) Host escape 1800 ft, from nest of Stegodyphus mimosa- with the prey is unlikely since A. ulova have rum on Zizyphus mucronata, 26, 1 OQ, 15 Dec. access to the hosts' entire trap-web and re- 1984, C. Griswold and T. Meikle-Griswold treat. Host retaliation is unlikely as Stego- (Natal Museum). dyphus are remarkably tolerant of other spi- 1987 GRISWOLD AND MEIKLE-GRISWOLD: ARCHAEODICTYNA ULO VA 9 ders. Various researchers have shown that creasing the number of spiders feeding on a inhibition of aggression in group-living Ste- prey item decreases an individual spider's en- godyphus species is not a property of colony ergy gain from that unit of prey. Possibly, A. membership or even restricted to members ulova has an adverse impact on its hosts, i.e., of the same species. By transferring individ- it increases feeding-group size without recip- uals between colonies several miles apart, rocally contributing to prey capture. Obser- Wickler (1973) showed that there was no vations ofA. ulova actively displacing Stego- within-colony recognition ofnest-mates in S. dyphus from prey items indicate that this mimosarum. We performed a similar exper- relationship may be detrimental to the hosts. iment using marked individuals. These were This impact might be minimized by the fact accepted into the new colonies, where they that A. ulova tends to join feeding groups participated in prey-capture and food-shar- during the later stages of prey consumption ing. We found this also to be the case with by the hosts. Laboratory observations showed S. dumicola. Additionally, we combined col- that A. ulova could even have a beneficial onies ofS. mimosarum and S. dumicola, and effect by causing trapped insects to struggle found that mixed colonies persisted in the in the web, attracting Stegodyphus to prey lab and on our verandah for several months. that might otherwise have been missed. Interactions during prey capture and feeding These observations raise a number ofques- appeared qualitatively the same as in single- tions concerning both the specific interac- species groups, and individuals touched one tions ofA. ulova with its Stegodyphus hosts, another frequently. This lack of species-level and more generally, the mechanisms and lim- discrimination demonstrates the potential its of tolerance in group-living spiders: Does ease of infiltration and integration of hetero- A. ulova cause a significant energy loss in its specifics. Kullman (1972) reported that intra- host species? Does it play a role in causing and interspecific tolerance, and inhibition of the hosts to abandon old nests? How do the biting in S. sarasinorum (another group-liv- dictynids locate host nests? What are the ing, cooperative species) are mediated by tac- physical and behavioral factors that allow A. tile and chemical cues on the integument. It ulova to integrate so well into Stegodyphus is possible that similarities in cuticular struc- nests? It is possible that highly integrated ture and/or semiochemicals aid in the inte- kleptobionts like A. ulova may provide useful gration ofA. ulova into Stegodyphus colonies. data for understanding the mechanisms of (7) Finally, A. ulova spiders move with ease tolerance and recognition in group-living spi- through the trap-web and interior tubes of ders. Stegodyphus nests. A. ulova is a cribellate spi- der, as are the Stegodyphus hosts. The hack- ACKNOWLEDGMENTS led-band silk of trap-webs and retreats con- structed by most members of the family We are grateful to the Natal Museum, Pie- Dictynidae are similar to those of the family termaritzburg, South Africa, for research fa- Eresidae, and this may be an advantageous cilities and partial support for fieldwork dur- preadaptation to infiltrating Stegodyphus ing this study. The Natal Parks, Fish, and nests. Game Preservation Board allowed use of While A. ulova apparently benefits in terms Spioenkop Dam Nature Reserve as a field of energy savings by not constructing trap- site, and generously provided accommoda- webbing or helping to catch prey, it is not yet tions. The National Parks Board, South Af- known whether the relationship between A. rica, provided research accommodations and ulova and its hosts is parasitic, commensal, field assistance at the Kruger National Park. or possibly even mutualistic. Ward and End- Support for this research was also provided ers (1985) showed that in groups of S. mim- by the Kalbfleisch Fund. For assistance in the osarum, increasing group size resulted in a field and collection of specimens, we thank decreased amount of overall food extraction Dr. R. Coville, Mrs. P. Coville, Mr. P. M. C. per spider per unit of time. They also found Croeser, Mrs. E. Griffin, and Mr. F. Khuna. a significant decline in individual biting times Scanning electron micrographs were made on as group size increased. This implies that in- a JEOL 2000 at the Electron Microscope Unit 10 AMERICAN MUSEUM NOVITATES NO. 2897 at the University ofNatal, Pietermaritzburg, Jackson, Robert R. and Ms. B. van Hoogdalem assisted in prep- 1978. Comparative studies of Dictyna and aration of illustrations. The scelionid wasps Mallos (Araneae, Dictynidae) I, social were identified by Drs. L. Masner and G. organization and web characteristics. Prinsloo. We thank Drs. J. A. Coddington, Rev. Arachnol., 1(4): 133-164. 1979. Comparative studies of Dictyna and R. R. Jackson, N. I. Platnick, and D. R. Smith, Mallos (Araneae, Dictynidae) II. The and Ms. M. Whitehouse for critically reading relationship between courtship, mating, drafts of the manuscript. aggression and cannibalism in species with differing types of social organiza- REFERENCES CITED tion. Ibid., 2(3): 103-132. 1985. The biology ofSimaetha paetula and S. Berland, Lucien thoracica, web-building jumping spi- 1914. Araneae (1st Part). In Voyage de Ch. ders (Araneae, Salticidae) from Queens- Alluaud et R. Jeannel en Afrique ori- land: cohabitation with social spiders, ental (1911-1912). Resultats scien- utilization of silk, predatory behaviour tifiques. Arachnida. Paris, vol.3, pp. 37- and intraspecific interactions. J. Zool. 94. London, ser. B, 1: 175-210. Berland, Lucien, and Jacques Millot Jackson, Robert R., and A. David Blest 1939. Les Araignees de I'Afrique occidentale 1982. The biology of Portia fimbriata, a web- Fran,caise, II. Cribellates. Ann. Soc. building (Araneae, Sal- Entomol. France, 108: 149-160. ticidae) from Queensland: utilization of Bradoo, Bhushan Lal webs and predatory versatility. J. Zool. 1979. Uloborus ferokus sp. nov. (Araneae: (London), 196: 255-293. Uloboridae), a commensal of Stegody- Kaston, Benjamin J. phus sarasinorum Karsch. Bull. Br. Ar- 1965. Some little known aspects of spider be- achnol. Soc., 4: 353-355. havior. Am. Midland Nat., 73: 336-356. Brockmann, H. Jane, and Christopher J. Barnard Kullman, Ernst J. 1979. Kleptoparasitism in birds. Anim. Be- 1972. Evolution of social behavior in spiders hav., 27: 487-514. (Araneae; Eresidae and Theridiidae). Cambridge, Octavius Pickard- Am. Zool., 12: 419-426. 1872. General list of the spiders of Palestine Lehtinen, Pekka T. and Syria, with descriptions of numer- 1967. Classification of the Cribellate spiders ous new species and characters of two and some allied families, with notes on new genera. Proc. Zool. Soc. London, the evolution of the suborder Araneo- 1872: 212-354. morpha. Ann. Zool. Fennici, 4: 199- 1876. Catalogue ofa collection ofspiders made 468. in Egypt, with descriptions of new Meikle-Griswold, Teresa C. species and characters of a new . 1986. Nest associates oftwo species ofgroup- Ibid., 1876: 541-630. living Eresidae in southern Africa. Ac- Caporiacco, Ludovico di tas X Congr. Int. Aracnol. Jaca/Espaiia, 1949. Aracnidi della colonia del Kenya rac- 1986, I: 275 [abstract]. colti da Toschi e Meneghetti negli anni Platnick, Norman I., and Mohammad U. Shadab 1944-1946. Commentat. Pontif. Acad. 1978. A review of the spider genus Mysme- Sci., 13(6): 309-492. nopsis (Araneae, Mysmenidae). Am. Denis, Jacques Mus. Novitates, 2661: 22 pp. 1954. Araignees recueillies par P. Remy du Simon, Eugene Sud Algerien au Hoggar. Bull. Soc. zool. 1899. Arachnides recueillis par M. C. J. De- France, 78: 311-324. witz en 1898, a Bir-Hooker (Wadi Na- Dewar, Margaret, and Harold Koopowitz tron), en Egypte. Bull. Soc. Entomol. 1970. Behavior of South African social spi- France, 13: 244-247. ders: prey capture. Am. Zool., 10: 476- Smith, Deborah R. 477 [abstract]. 1982. Reproductive success of solitary and Griswold, Charles E. communal Philoponella oweni (Ara- 1985. okuncana, a newgenus and species neae: Uloboridae). Behav. Ecol. Socio- of kleptoparasitic spider from southern biol., 11: 149-154. Africa (Araneae: Mysmenidae). Ann. Tullgren, Albert Natal Mus., 27: 207-217. 1910. Araneae. In Wissenschaftliche Ergeb- 1987 GRISWOLD AND MEIKLE-GRISWOLD: ARCHAEODICTYNA ULOVA 11

nisse der Schwedischen Zoologischen Ward, Paul I., and Margit M. Enders Expedition nach dem Kilimandjaro, 1985. Conflict and cooperation in the group dem Meru und dem Umgebenden Mas- feeding ofthe Stegodyphus saisteppen Duetsch-Ostafrikas 1905- mimosarum. Behaviour, 94: 167-182. 1906 unter Leitung von Prof. Dr. Yngve Whitehouse, Mary Sjostedt, vol. 20. Stockholm: Palm- 1986. The foraging behaviours of Argyrodes quists, pp. 85-172, pls. 1-4. antipodiana (Theridiidae), a kleptopara- Vollrath, Fritz sitic spider from New Zealand. New 1979. A close relationship between two spi- Zealand J. Zool., 13: 151-168. ders (Arachnida, Araneidae): Curima- Wickler, Wolfgang gua bayano synecious on a Diplura 1973. Uber Koloniegrundung und soziale Bin- species. Psyche, 85(4): 347-353. dung von Stegodyphus mimosarum Pa- 1984. Kleptobiotic interactions in inverte- vesi und anderen sozialen Spinnen. Z. brates. In C. J. Barnard (ed.), Producers Tierpsychol., 32: 522-531. and scroungers: strategies of exploita- Wilson, Edward 0. tion and parasitism. London and Syd- 1971. The insect societies. Cambridge: Har- ney: Croom Helm, pp. 61-94. vard Univ. Press. 1987. Kleptobiosis in spiders. In W. Nentwig (ed.), Ecophysiology of spiders. New York: Springer-Verlag, pp. 274-286. Recent issues of the Novitates may be purchased from the Museum. Lists of back issues of the Novitates, Bulletin, and Anthropological Papers published during the last five years are available free of charge. Address orders to: American Museum of Natural History Library, Department D, Central Park West at 79th St., New York, N.Y. 10024.