: PHYLOGENY, SOCIALITY AND INTERSPECIFIC INTERACTIONS—A REPORT ON THE ARGYRODES SYMPOSIUM, BADPLAAS 2001 Author(s): Mary Whitehouse, Ingi Agnarsson, Tadashi Miyashita, Deborah Smith, Karen Cangialosi, Toshiya Masumoto, Daiqin Li, and Yann Henaut Source: Journal of Arachnology, 30(2):238-245. Published By: American Arachnological Society DOI: http://dx.doi.org/10.1636/0161-8202(2002)030[0238:APSAII]2.0.CO;2 URL: http://www.bioone.org/doi/full/10.1636/0161-8202%282002%29030%5B0238%3AAPSAII %5D2.0.CO%3B2

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. 2002. The Journal of Arachnology 30:238±245

ARGYRODES: PHYLOGENY, SOCIALITY AND INTERSPECIFIC INTERACTIONSÐA REPORT ON THE ARGYRODES SYMPOSIUM, BADPLAAS 2001

Mary Whitehouse1: Mitrani Center, Jacob Blaustein Institute, Ben Gurion University, 84990 Israel and Department of Zoology and Entomology, The University of , Brisbane Q1d 4072 Ingi Agnarsson: Department of Biological Sciences, George Washington University, 2023 G Street NW, Washington, D.C. 20052 USA Tadashi Miyashita: Laboratory of Biological Sciences, School of Agriculture and Life Science, University of Tokyo, Tokyo 113-8657 Deborah Smith: Entomology Program, Department of Ecology and Evolutionary Biology, University of Kansas, Lawrence, KS 66045-7534 USA Karen Cangialosi: Biology Department, Keene State University, Keene, NH 03435- 2001 USA Toshiya Masumoto: Center for Ecological Research, Kyoto University Kamitanakami, Otsu, Shiga 520-2113 Japan Daiqin Li: Department of Biological Sciences, National University of , 119260 Singapore Yann Henaut: Laboratorio EcoetologõÂa de ArtroÂpodos, El Colegio de la Frontera Sur, Tapachula, Chiapas,

ABSTRACT. Argyrodes Simon 1864 is a large, cosmopolitan theridiid whose members exhibit a wide range of foraging techniques which usually involve exploiting other , either by using their webs, stealing their food, or preying on them directly. We held a symposium on this genus at the 15th International Congress of Arachnology, Badplaas, South in order to obtain a clearer perspective on the relationship between the phylogeny of the genus and the different foraging techniques. We concluded that Argyrodes forms a monophyletic group within the , and that there are clear monophyletic clades within the genus (already identi®ed as species groups) that appear to share behavioral characteristics. We found no clear indication that foraging behaviors such as kleptoparasitism (stealing food) evolved from araneophagy (eating spiders) or vice versa. However, it appears that species that specialize in either kleptoparasitism or araneophagy use additional techniques in comparison to species that readily use both foraging modes. During our examination of Argyrodes/host interactions we noted the importance of Ne- phila species as hosts of Argyrodes species around the world and the impact of Argyrodes on . We also noted the ¯uid nature of the relationship between Argyrodes and the spiders with which they interact. For example, an Argyrodes/host relationship can change to an Argyrodes/prey relationship, and the type of kleptoparasitic behavior employed by an Argyrodes can change when it changes host species. The importance of eating silk was also noted and identi®ed as an area for further research. We concluded that more work involving international collaboration is needed to fully understand the phylogeny of the genus and the relationships between the different types of foraging behaviors.

The large (over 200 species) cosmopolitan many of its species and their unusual foraging genus Argyrodes has attracted interest techniques (which include invading webs to worldwide because of the gregarious nature of steal food from and to attack other spiders). In response to increasing international attention in 1 Current address: CSIRO Entomology, Australian this group we decided to hold a symposium on Cotton Research Institute, Locked Bag 59, Narrabri, Argyrodes to consolidate our knowledge and NSW 2390, Australia. obtain an overall perspective on the genus. Our

238 WHITEHOUSE ET AL.ÐARGYRODES SYMPOSIUM REPORT 239

Figure 1.ÐChart of the six recognized species groups of Argyrodes (from the Americas) indicating the current known foraging behaviors and the standard form of the male cephalothorax for each group. A tick indicates that a species from that group performs the foraging technique, a cross indicates that a species is known not to perform this behavior, and a question mark indicates that nothing is known about the foraging method in relation to the species group. A tick and a cross for the same foraging method indicates that some species in this group use the foraging method while other species do not. aim was to identify the direction that the re- foraging techniques. Below is a report on the search was leading, and develop future re- conclusions we drew from the symposium and search programs that are more integrated. the areas that still require more research. The symposium was loosely focused on un- Evolution.ÐEvolutionary relationships derstanding how the phylogenetic relation- within the genus Argyrodes are poorly under- ships within the genus re¯ected the evolution stood. Currently there are six recognized spe- of different types of relationships with other cies groups (Exline & Levi 1962) within the spider species. The ``interaraneae'' relation- genus: Argyrodes, , , ships of Argyrodes species are very diverse. Cordillera, Cancellatus, and Trigonum. Be- Some species behave as kleptoparasites in that cause these names refer to species groups and they invade the web of a (usually larger) host currently not to genera, they are not in italics. spider and eat the host's web, glean insects off It is confusing that ``Argyrodes'' refers to the the host's web, steal the host's wrapped food whole genus and to a particular species group. bundles, and/or feed with the host. Some spe- In this text, when we refer to the genus Ar- cies attack the host when it is vulnerable such gyrodes we will use italics, but when we refer as during molting, or capture and eat small to the species group Argyrodes, we will use spiders by lunging at them and grabbing them normal script. The evidence to date suggests with their front legs. Still others capture spi- that in the species groups may use ders by throwing a line of sticky silk over the similar methods of web invasion (Fig. 1). For victim. Phylogeny provided a framework in example, all species so far studied in the which we discussed 1) these diverse interspe- Rhomphaea and Ariamnes species groups ci®c interactions, 2) sociality, and 3) speci®c seem to specialize on araneophagy (Eberhard 240 THE JOURNAL OF ARACHNOLOGY

Figure 2.ÐFour ways in which the evolution of araneophagy and kleptoparasitism may be related. See text for details of each model.

1979; Horton 1982; Whitehouse 1987). They only glean insects and eat the host's silk (A. tend to be free-living and solitary, and capture caudatus Taczanowski 1874: Henaut & Ibar- spiders by throwing a sticky silk line over the ra-Nunez unpubl. data; Vollrath 1984) and victim. Species from the Argyrodes species other members that will also feed with the group are gregarious and seem to specialize host (A. globosus Keyserling 1884: Henaut on kleptoparasitism. They will even feed with 2000) and others which will not feed with the the host to obtain food: A. antipodianus O.P. host, but will steal food bundles (A. ululans Cambridge 1880 (Whitehouse 1986, 1997; O.P. Cambridge 1880: Cangialosi 1990a, b). Grostal & Walter 1997); A. elevatus Tacza- Thus in the Cancellatus species group there is nowski 1872 (Vollrath 1979, 1984); A. argen- no consistency in the kleptoparasitic tech- tatus O.P. Cambridge 1880 (Robinson & Rob- niques used. No spiders from the Cordillera inson 1973); A. argyrodes 1842 (Kullmann species group have been studied. 1959). Species that have been studied from Four pathways have been proposed by the Trigonum species group forage using both which kleptoparasitism and free-living ara- kleptoparasitism and araneophagy: A. trigo- neophagy may have evolved (Fig. 2). First, num Hentz 1850 (Cangialosi 1997; Larcher & ecological pressures, rather than evolutionary Wise 1985; Suter, et al 1989) and A. babo- history, may have dictated which behavior is quivari Exline & Levi 1962 (Larcher & Wise expressed in each species so that there is no 1985). However, the araneophagy that A. tri- phylogenetic relationship between phylogeny gonum (at least) uses is distinct from that of and behavior (Model 1). Alternatively, ara- Rhomphaea and Ariamnes species. Cangialosi neophagy and kleptoparasitism may each have reported in the symposium that it does not evolved once, in which case there are three throw silk in order to capture the spider, but possible models: Free-living araneophagy kills the spider by biting it. The species group may have evolved from kleptoparasitism Cancellatus contains some members that will (Model 2). Smith Trail (1980) argued that the WHITEHOUSE ET AL.ÐARGYRODES SYMPOSIUM REPORT 241 kleptoparasitic skills of interpreting the host's neophagic behavior of ``throwing a sticky vibrations could preadapt Argyrodes for safely thread over the prey''). The behavior of the stalking and capturing the host itself. Alter- basal species of both Agnarsson's and Masu- natively, kleptoparasitism may have evolved moto's trees directly ®t this model, while the from araneophagy (Model 3). Vollrath (1984) behavior of the basal species in Whitehouse's supported this model although he argued that tree is not known, except that it occurs on a Argyrodes would initially invade other spi- larger spider's web. Consequently, the avail- ders' webs and chase out the owner, and then able evidence to date suggests that the araneo- later adopt araneophagic behaviors that would phagic and kleptoparasitic foraging behaviors preadapt them to kleptoparasitism. Finally, of Argyrodes species evolved concurrently, and both kleptoparasitism and araneophagy may latter species may have specialized, and/or re- have evolved separately (Model 4). White- ®ned these techniques. house (1987) proposed this argument based on In addition, all three trees support some differences in the araneophagic techniques of general claims. For example, the trees of both species from the predominantly araneophagic Agnarsson and Whitehouse indicate that Ar- (Rhomphaea and Ariamnes) and kleptoparas- gyrodes and Ariamnes are sister species itic (Argyrodes) species groups. groups, while all three trees suggest that the The three phylogenetic studies presented at Rhomphaea species group is quite distinctive. this symposium examined the relationship be- This suggests that Rhomphaea and Ariamnes tween these different species groups and their may have developed araneophagic foraging foraging techniques, in particular the relation- techniques independently of each other. ship between species that are predominantly It is intriguing that Rhomphaea and Ariam- kleptoparasitic, and those that are predomi- nes may have developed araneophagy indepen- nantly araneophagic (Fig. 3). Agnarsson pre- dently because the technique they both use to sented a phylogenetic tree of Argyrodes (large- capture spiders (throwing silk) is distinctive ly from the Americas) within the context of the from the technique used by the basal Argyrodes family Theridiidae, and used sequences from species (biting/lunging). At the symposium we the genes CO1, 16S, 18S and 28S and mor- debated whether the spider-catching behavior phological characters to construct the tree. of Ariamnes and Rhomphaea was plesiom- Masumoto, working on Japanese species, con- orphic or derived. Most symposium members structed his tree using sequences from the gene (who have not seen Ariamnes or Rhomphaea CO1, while Whitehouse presented trees of catch spiders) regarded it as a plesiomorphic Australian Argyrodes based on sequences from theridiid trait because most theridiids catch the genes CO1 and 16S (for more information prey by wrapping them with sticky silk. Whi- see Agnarsson et al this journal, Masumoto un- tehouse argued that the behavior is derived be- publ. data, Whitehouse et al. unpubl. data). cause it is very distinctive from normal theri- Super®cially, all three trees appear to sup- diid wrapping. Theridiids normally attack prey port different models: Agnarsson's tree seems by throwing numerous threads of silk in quick to support model 1 (no evolutionary relation- succession with alternating legs IV over the ship between developing araneophagy and victim until it is completely covered. When kleptoparasitism); Masumoto's tree seems to Rhomphaea and Ariamnes attack prey, the two support model 2 (araneophagy developed from legs IV move in unison towards the prey, and kleptoparasitism) and Whitehouse's tree seems the spider will throw one to ®ve sticky threads. to support model 3 (kleptoparasitism developed Once the prey is immobilized Rhomphaea/Ar- from araneophagy). However Bremer support iamnes will assume normal theridiid wrapping for the lower nodes are not strong in any tree, behavior. Whitehouse conceded that within and relationships between species could easily these two species groups there might be a con- switch around. In addition the trees suggest that tinuum in that some species may throw silk the basal species within the genus Argyrodes more like a standard theridiid while others may use both kleptoparasitic and araneophagic be- be more distinctive and more stylized. Agnars- haviors, even though they lack the more de- son suggested that a solution would be to look rived techniques of these foraging methods for the spigots on the spinnerets that are re- (such as the more derived kleptoparasitic be- sponsible for producing sticky silk in theri- havior of ``feeding with the host'', or the ara- diids. He noted that individuals in the Argy- 242 THE JOURNAL OF ARACHNOLOGY WHITEHOUSE ET AL.ÐARGYRODES SYMPOSIUM REPORT 243 rodes (``kleptoparasitic'') species group have of webs. Because of the size of the group and lost one of the two aggregates on each PLS. the morphological diversity within the group it Miyashita added that kleptoparasitic Argyrodes is possible that Cancellatus is not monophylet- also lack an aggregate gland for producing ic. A comprehensive phylogeny would reveal sticky silk. If basal species do not have these this. Nevertheless we need to show caution spigots and an aggregate gland, then this would when deciding which species will be represen- suggest that spider-catching method used by tative of species groups. Rhomphaea and Ariamnes species is derived. Sociality.ÐAn interesting aspect of the If basal species do have these structures, then theridiid phylogeny that Agnarsson pointed ``throwing silk'' is more likely to be a plesio- out and which he discusses in this volume morphic trait. (Agnarsson et al 2001) was that Argyrodes Our discussions on the phylogeny of Ar- form a monophyletic clade with the genera gyrodes emphasized the need for more infor- that contain social spiders. One of the striking mation. Firstly we need a more comprehen- characteristics of many species of Argyrodes sive phylogenetic tree to identify all species is that they are gregarious, even forming groups. We concluded that currently named mixed species groups around other spider's species groups (Argyrodes, Rhomphaea, Ar- webs. Their location within the theridiid phy- iamnes and Trigonum) appear to be monophy- logeny suggests that they may have a phylo- letic and therefore useful groupings of the spe- genetic predisposition to form groups. cies. However these species groups are only The signi®cance of the group-forming be- speci®c for American species, and that species havior may be that it enhances the effective- in other continents, like Asia, Australia and ness of kleptoparasitism. For example, many Africa, may form different species groups. We Argyrodes on the same host's web will be pro- concluded that we need an integrated, com- ducing vibratory signals from numerous di- prehensive phylogenetic tree that includes rections, confusing the host. Henaut pointed species found throughout the world, to estab- out that distraction had the effect of cooper- lish if species groups within the Argyrodes ation. He observed A. globosus distract the complex are indeed monophyletic and should host while another A. globosus stole the food. be recognized as separate genera. He also saw A. globosus vary its degree of Secondly, we acknowledged that there is a gregariousnessÐit was more gregarious on huge lack of behavioral data, and that it is un- the webs of the more aggressive host (Leu- likely that we can obtain behavioral data from cauge mariana Taczanowski 1881, L. venusta each of the 200 species worldwide. We con- Walckenaer 1842 and L. argyra Walckenaer cluded that a better approach would be to iden- 1842) than the less aggressive host (Gaster- tify the monophyletic species groups within the ancantha cancriformis (Linnaeus 1758)). genus and then focus on particular species Host-Argyrodes interactions.ÐAnother im- within these groups. Henaut expressed caution portant theme in the symposium was the rela- with this approach. His point was well taken tionship between hosts and Argyrodes. Firstly, as the large Cancellatus species group is known Miyashita looked at the effect of different types to contain a species (A. globosus Henaut 2000) of host species on the distribution of Argyrodes that can do a range of kleptoparasitic tech- in Japan. He found that Argyrodes were limited niques including feeding with the host, while by the distribution of their hosts and that Ne- it also contains a species (A. caudatus) which phila spp. were particularly important. Li also has been studied intensively (Vollrath 1984, pointed out the strong relationship between Ar- Henaut & Ibarra-Nunez unpubl. data) but gyrodes and Nephila in Singapore, and this re- which only gleans insects from around the edge lationship has also been noted in the Americas

←

Figure 3.ÐPhylogenetic trees of species groups within Argyrodes reported by Agnarsson, Masumoto and Whitehouse at the symposium, indicating foraging behaviors associated with the species groups. Only spiders from the Americas (Agnarsson's tree) have been formally assigned to the different species groups, so the speech-marks indicate the most probable species groups for the spiders from Japan (Masumoto) and Australia (Whitehouse). 244 THE JOURNAL OF ARACHNOLOGY

(Vollrath 1979) and in Australia (Elgar 1989, between A. trigonum and its host (from klep- 1993; Grostal & Walter 1997). As Smith noted, toparasite/host to predator/prey) provides an- the importance of Nephila spp. as a host spe- other dimension to the ``interaraneae'' interac- cies for Argyrodes appears to be pandemic. tions within the genus Argyrodes. The Secondly, the actual relationship between ¯exibility of the ability to change and the eco- host and Argyrodes was explored. Li and Can- logical rami®cations of the change for both the gialosi emphasized that species of Argyrodes host and the kleptoparasite were discussed and are often assumed to be kleptoparasitic (i.e. de- seen as important areas for future development. rogatory to the welfare of the host) when they Many questions remain concerning the fac- could be commensal (have no effect on the tors contributing to changes in the relationship host). Li provided evidence that A. ¯avescens between Argyrodes species and their hosts. O. P. Cambridge 1880 did have a direct affect Basal species, such as those from the Trigo- on its host Nephila pilipes (Fabricius 1793). In num species group which exhibit a wide rep- the presence of the kleptoparasite, N. pilipes ertoire of foraging behaviors, not only provide were smaller and produced fewer, but larger insight into the evolution of kleptoparasitism eggs. We concluded that the effect of Argyro- and araneophagy, but (as both Miyashita and des on the ®tness of the host was an area that Cangialosi pointed out) can also be good could be expanded. models for studying how shifting behavioral Thirdly, the relationship between the host relationships between species can translate and the Argyrodes can change depending on into complex patterns of population dynamics. the type of host, and even the developmental More subtle changes in the relationship be- stage of the Argyrodes. Cangialosi, working tween Argyrodes and their hosts were also with a phylogenetically basal species A. trigo- presented at the symposium. Henaut showed num, demonstrated that while this species ex- that A. globosus would only use the behavior hibits both araneophagy and kleptoparasitism ``feeding with the host'' with the less aggres- for all three hosts that she has studied; it is sive hosts (Gasterancantha cancriformis, Ver- predominately a predator of Neriene radiata rucosa arenata (Walckenaer 1842), and Ne- (Walckenaer 1842) (Linyphiidae) and predom- phila clavipes (Linnaeus 1767)) and would inately a kleptoparasite of Pityohyphantes cos- form larger groups around the webs of the tatus (Hentz 1850) (Linyphiidae) and Achaear- more aggressive host ( mariana, L. anea tepidariorum (C. L. Koch 1841) venusta and L. argyra) that had the effect of (Theriididae). distracting the host. Whether A. trigonum behaves as a klepto- An unusual relationship highlighted by parasite has to do not only with relative host Smith was that between an unnamed species size, but also with the developmental stage of of Argyrodes and plants protected by ants Argyrodes independent of its relative size. For (Fowler & Venticinque 1996). In this case Ar- example, older A. trigonum are more likely to gyrodes is not interacting with other spider be aggressive compared to juveniles, regardless species but with ants. How the Argyrodes in- of host size. teracts with the ants, and how this species of Although A. trigonum switched between Argyrodes relates phylogenetically to other kleptoparasitism and araneophagy, the behav- Argyrodes species, are two additional areas of ioral repertoire within each of these categories research that need developing. was limited. Kleptoparasitically, Cangialosi re- Silk eating.ÐBoth Miyashita and Smith ported that A. trigonum gleaned insects and emphasized the importance of Argyrodes con- stole prey, but that it did not feed with its host suming the silk of its host. Miyashita pointed (a kleptoparasitic behavior common in the Ar- out that this behavior enables Argyrodes to sur- gyrodes species group) or eat silk. Araneo- vive periods of low prey abundance in the phagically, Cangialosi reported that A. trigo- host's web. It would be interesting to know num attacked spiders by biting them, but that how widespread this behavior is (Cangialosi it did not throw a silk line over a prey spider reported that she has not seen A. trigonum feed in order to catch it (the araneophagic method on silk despite many hours of observations). of species in the Rhomphaea and Ariamnes Many species of spiders eat their own silk; do species groups). many species eat other spider's silk as well? Nevertheless the change in the relationship This area also needs further investigation. WHITEHOUSE ET AL.ÐARGYRODES SYMPOSIUM REPORT 245

Conclusions.ÐResearch within the genus tualism. Memoirs of the Queensland Museum Argyrodes is at a very interesting stage. Our 33:411±430. ®rst priority is to improve our understanding of Exline, H. & Levi, H.W. 1962. American spiders of the phylogeny and its relationship to the mul- the genus Argyrodes. Bulletin of the Museum of titude of foraging techniques common within Comparative Zoology 127:75±203. Fowler, H.G. & Venticinque, E.M. 1996. Spiders the genus. With these points clari®ed we can and understory myrmecophytes of the central more easily address ecological questions con- Amazon, . Revista brasileira de entomolo- cerning interspeci®c interactions between Ar- gia 40:71±73. gyrodes and their ``hosts''. Our results suggest Grostal, P. & Walter, D.E. 1997. Kleptoparasites or that different ecological questions may be par- commensals? Effects of Argryodes antipodianus ticularly relevant for different species groups. (Araneae: Theridiidae) on Nephila plumipes (Ar- For example, members of the Trigonum species aneae: Tetragnathidae). Oecologia 111:570±574. group may be particularly useful for investi- Henaut, Y. 2000. Host selection by a kleptoparasitic gating a switch from kleptoparasitic behaviors spider. Journal of Natural History 34:747±753. to predatory behaviors. Species in the Argy- Horton, C.C. 1982. Predators of two orb-web spi- ders (Araneae, Araneidae). Journal of Arachnol- rodes species group may be useful when asking ogy 11:47. questions either about host speci®city or con- Kullmann, E.J. 1959. Beobachtungen und Betrach- ditions under which an Argyrodes should tungen zum verhalten der Theridiidae Conopis- change its kleptoparasitic techniques. tha argyrodes Walckenaer (Araneae). Mitteilun- Obviously, these topics are only the tip of gen 35:275±292. the iceberg. Our symposium only touched on Larcher, S.F. & Wise, D.H. 1985. Experimental the question of Argyrodes and sociality, the studies of the interactions between a web-invad- role of crypsis in determining the striking ing spider and two host species. Journal of Ar- morphology of many Argyrodes species, and achnology 13:43±59. Robinson, M.H. & Robinson, B. 1973. Ecology and how Argyrodes locate their hosts. Mixed spe- behavior of the giant wood spider Nephila ma- cies groups of Argyrodes were not discussed, culata (Fabricius) in New Guinea. Smithsonian and we did not mention courtship behavior at Contributions to Zoology 149:1±70. all. All this indicates a very exciting and in- Smith Trail, D. 1980. by Argyrodes teresting future for behavioral-ecology re- (Theridiidae) on solitary and communal spiders. search within the genus Argyrodes. Psyche 87:349±355. Suter, R.B., Shane, C.M. & Hirscheiner, A.J. 1989. LITERATURE CITED Spider vs spider: Frontinella pyramitela detects Argyrodes trigonum via cuticular chemicals. Agnarsson, I., Arnedo, M.A., Gillespie, R., Cod- Journal of Arachnology 17:237±240. dington, J.A. & Hormiga, G. 2002. Sharing a Vollrath, F. 1979. Behavior of the kleptoparasitic web: a phylogenetic view of sociality and klep- spider (Araneae, Theridi- toparasitism in cobweb spiders (Araneae, Theri- idae). Behaviour 27:515±521. diidae). Journal of Arachnology 30:pp. Vollrath, R. 1984. Kleptobiotic interactions in in- Cangialosi, K.R. 1990a. Life cycle and behavior of vertebrates. In: Producers and Scavengers: strat- the kleptoparasitic spider, Argyrodes ululans. egies of exploitation and parasitism. (ed. by C. Journal of Arachnology 18:347±358. J. Barnard), pp. 61±94. London: Grom Helm. Cangialosi, K.R. 1990b. Social spider defense Whitehouse, M.E.A. 1986. The foraging behaviors against kleptoparasitism. Behavioural Ecology of Argyrodes antipodiana (Araneae, Theridi- and Sociobiology 27:49±54. idae), a kleptoparasitic spider from . Cangialosi, K.R. 1997. Foraging versatility and the New Zealand Journal of Zoology 13:151±168. in¯uence of host availability in Argyrodes tri- Whitehouse, M.E.A. 1987. Spider eat spider: The gonum (Araneae, Theridiidae). Journal of Arach- predatory behavior of Rhomphaea sp. indet. from nology 25:182±193. New Zealand. Journal of Arachnology 15:355±362. Eberhard, W.G. 1979. Argyrodes attenuatus (Ther- Whitehouse, M.E.A. 1997. The bene®ts of stealing idiidae): a web that is not a snare. Psyche 86: from a predator: Foraging rates, predation risk 407±413. and intraspeci®c aggression in the kleptoparasitic Elgar, M.A. 1989. Kleptoparasitism: a cost of ag- spider Argyrodes antipodiana. Behavioral Ecol- gregating for an orb-weaving spider. Animal Be- ogy 8:663±667. haviour 37:1052±1055. Elgar, M.A. 1993. Inter-speci®c associations involv- Manuscript received 13 August 2001, revised 15 ing spiders: kleptoparasitism, mimicry and mu- March 2002.