Fonuu How Make a Living

M. NYFFELER, W. L. STERLING, aNo D. A. DEAN

Department of Entomology, Texas A&M University, College Station, TX77843

Environ. Entomol. 23(6): 1357-1367 (1994) ABSTRACT Although the beneffcial status of the spiders as insectivores has been widely recognized for quite some time, biologists by and large seem to be rather unfamiliar with the speciffc feeding habits of this very diverse order. We present an overview of the feeding patterns of 10 groups of common agroecosystem spiders to inform entomologists and ecologists concerned with issues of natural biological control. The various groups discussed in this article exhibit a very diverse range of life styles and foraging modes, which is reflected in the diversity of their feeding patterns. Implications of the insectivorous activities ofthese predators for natural pest control are discussed.

KEY WORDS spiders, predation, diets

Sprorns (AneNner) ARE a very diverse order of ingest solid food and must, therefore, inject di- ubiquitous carnivores within the class Arach- gestive enzymes into the immobilized prey (ex- nida. At the present time, >30,000 species of ternal digestion) and then suck in the dissolved spiders are described (Coddington & Levi 1991). tissue in liquid form. Spiders generally have a Over 3,000 species occur in North America alone very low rate of metabolism compared with other (Young & Edwards 1990). The vast majority of poikilothermic organisms of comparable body spiders occupy terrestrial habitats. Some ly- weight (Greenstone & Bennett 1980). They can cosids and pisaurids, however, can walk and sail store energy and starve for considerable time on the water surface (and at times even dive and periods, which makes them excellent survivors swim under water); they forage on aquatic and under conditions of food shortage (see Nyffeler & semiaquatic organisms when they inhabit Breene I990a). marshes, flooded rice fields, and other wetlands According to traditional foraging theory, spi- (Greenstone 1979, Oraze & Grigarick 1989, Zim- ders are considered to be predators of live, mov- merrnann & Spence 1989). One agelenid species ing prey only (e.g., Turnbull 1960, 1973). More (the water spider) actually lives under water (see recent studies have modiffed this view when ev- subsection Foraging Patterns of Web Weaoers). idence was found that spiders utilize a much Most spiders are highly cannibalistic solitary broader range of foraging strategies, including creatures and practice bizarre courtship rituals feeding on eggs (oophagy), dead ani- (Tumbull 1973). Several species produce sounds mals (scavenging), plant pollen, and even ar- (acoustic communication) during courtship and tiffcial diets (see McDaniel & Sterling 1982, agonistic displays (Rovner 1975,Uetz & Stratton Nyffeler et al. 1990a). Stealing ofprey from other 1982). These live in a world full of vi- spiders (kleptoparasitism) plays an important brations (e.g., Rovner & Barth l98l). Sexual di- role as an alternative foraging strategy ofvarious morphism occurs in many species, the female web spinners (Vollrath f987). Spiders have been normally being significantly larger than the male reported feeding on a wide range of different (hereafter adult length always refers to the fe- groups including some unusual prey such male). Spiders disperse by walking on the as small mice, bats, birds, ffsh, crayfish, crabs, ground, by using silk-thread bridges between frogs, lizards, snakes, and scorpions (Nyffeler & plants, as well as ballooning through the atmo- Benz 1981, McCormick & Polis 1982); however, sphere from place to place on silken threads in general they tend to concentrate on insect (Foelix 1982, Dean & Sterling 1985, Young & prey and to a lesser degree on other spiders Edwards 1990). All spiders produce silk from (Wise 1993). Most spider species forage on mul- abdominal glands though only the useb weaoers tiple prey species (generalist predators), which construct webs that are used to catch prey. Spi- Greenstone (1979) has suggested may be advan- ders are equipped with a pair ofjaws (chelicerae) tageous by optimizing a balanced essential and possess venom glands (exception, Ul- amino acid composition in the diet. Spiders feed oboridae do not produce venom). Immobiliza- predominantly on small-sized prey relative to tion of prey is assisted by the use of silk and by their own size (prey length s predator length) the injection of venom. These animals cannot (Nyffeler & Benz 1981, Wise 1993); feeding ex-

OO46-225X/941 1357-1367$02.00/0 @ 1994 Entomological Socie9 of America This article is the copyright property of the Entomological Society of America and may not be used for aqy commercial or other private purpose without specific r,rnitten permission of the Entomological Society of America r358 ENvrnoNurrvrer, EurouolocY Vol.23, no.6

periments with a variety of spider species and a Swiss Federal Institute of Technology, respec- model prey (crickets) conducted in the labora- tively. tory revealed that the optimal prey length ranges from 50-80Vo of the spider length (Nentwig re87). Results and Discussion Spiders are among the numerically dominant Foraging Patterns of Web Weavers. High feed- insectivores in terrestrial ecosystems and exhibit ing frequencies (up to 90Vo spiders feeding si- a vety diverse range of life styles and foraging multaneously during peak activity) were ob- behaviors (Turnbull 1973, Wise 1993). Two basic served in ffeld populations ofcertain larger-sized groups of foraging strategies can be distin- orb weavers (Araneidae) that rebuild (recycle) guished: (I) web spöders (i.e., foraging with a their webs daily (Nyffeler 1982). The high feed- catching web) (Tables l-5), and (2) hunters or ing frequencies indicate that the web is a very wanderers (i.e., foraging without the use of a efficient prey capturing device. Large orb weav- web) (Tables 6-9). Some prominent representa- ers often kill prey in excess of their energy re- tives of web spinning spiders are orb weavers quirements. As many as 1,000 small insects have (Araneidae and Tetragnathidae), sheet web been found entangled at one time in a single orb weavers (Linyphiidae), mesh web weavers (Dic- web; however, not all insects caught by the web tynidae), comb-footed spiders (Theridiidae), and are eaten. Sheet-web weavers, mesh web weav- funnel-web weavers (Agelenidae). Prominent ers, comb-footed spiders, and funnel-web weav- representatives of hunters are wolf spiders (Ly- ers that do not renew their nets daily, feed cosidae), lynx spiders (Oxyopidae), crab spiders less frequently (

Table l. Prey spectrum (in percent) of small orb- phus, and Schistocerca) constituted 187o of the weaver, Tetragnatha laborioto llentz (Ietragnathidae), total prey intercepted in the cm diameter baaed on three different field studiee -30 webs ofArgiope aurantia Lucas in a cotton ffeld (Table Prey type Study l" Study 2b Study 3" of East Texas 2); this spider (:20-25 mm adult length) kills prey up to :2O0Vo of its own t7.5 t2.2 Diptera 40.5 size (Nyffeler et 1987a). Cicadellidae 36.7 50.0 /.J al. Aphididae 0.0 12.5 78.0 Orb weavers generally seem to be rather inef- Other Homoptera 1.3 /.ö 0.0 fective in trapping moths and butterflies. Eisner Heteroptera 17.7 2.5 0.0 et al. (f 964) stated: "Moths, by virtue of the loose Coleoptera 0.0 5.0 0.0 cover wings and bodies, are Fomicidae 2,6 2.5 2.5 scales that their Lepidoptera 0.0 2.5 0.0 admirably adapted to elude capture by orb- Others 1.2 0.0 0.0 weaving spiders. Rather than sticking to the web, Total r00.0 100.0 100.0 they may simply lose some of their scales to the No. prey records 79 40 4t viscid threads, and fly on." Nyffeler (1982) re- corded that flying lepidopterans made up a very " In soybeans in Illinois (LeSar & Unzicker 1978). low percentage of the prey of various temperate b In soybeans in Kentucky (Culin & Yeargan 1982). 'In cotton in Texas (Nyffeler et al. 1989). orb weaver species. Several species oforb weav- ers, however, spin highly modiffed orb webs (e.g., Iadder web and bolas spi,der) that function ing entangled beetles from the web by the fol- as effective moth traps (Foelix 1982, Eberhard lowing tactics: (l) by the spider violently shaking r99o). the web until the beetle fell, (2) by ignoring the Sheet-Web Weaoers. The family of sheet- beetle until it worked itself free and could es- web weavers (Linyphiidae) includes the subfam- cape, (3) by cutting the web around an adult ilies Linyphiidae-Linyphiinae and Linyphiidae- beetle allowing it to drop from the web (LeSar & Erigoninae (:Erigonidae or Micryphantidae). Unzicker 1978, Culin & Yeargan 1982). Fragile, These spiders hang inverted below the sheet small nets of small orb weavers such as T. labo- waiting for prey, which they pull through the röosa are suitable for interception of small insects sheet (Wise 1993). Linyphiid webs include some only (narrow feeding niche) (LeSar & Unzicker viscid silk though it does not seem to be much 1978, Culin & Yeargan 1982). involved in prey capture. Various small to medi- In contrast, large orb weavers of the family um-sized species of the subfamily Linyphiinae Araneidae are able to overcome the defenses of a can reach high abundance in woodlands and wider diversity of prey types, with their strong grasslands where they kill numerous small in- nets (broad feeding niche), which include in- sects primarily from the orders Diptera, Hyme- sects with strong sclerotization, chemical protec- noptera, Homoptera, and Heteroptera (Turnbull tion, and aggressive behavior (Culin & Yeargan 1960, Nyffeler & Benz 1981). Lepidopterans and 1982, Nentwig f987). Large orb weavers of the coleopterans often escape from the fragile sheet genus Argiope frequently kill grasshoppers (Or- webs and, thus, compose an insignificant fraction thoptera) and large stinging bees (including Apis of these spiders' diet (Turnbull 1960). Dwarf spi- mellifera L.) (Table 2; Nyffeler & Breene l99l). ders of the subfamily Erigoninae (Erigone spp. Grasshoppers (genera Melanoplus, Encoptolo- and Oedothorax spp.), (3 mm in length, numer- ically dominate the spider faunas on the ground Table 2. Prey spectrum (in percent) of large orb- surface of agricultural ffelds in the temperate- weavem (Araneidae) baeed on three different field studiee northern zones (Sunderland et al. 1986, Nyffeler & Benz 1988a). With fragile small sheet webs Prey type Study 1" study 2b Study 3' spun horizontally over small depressions on the Diptera 26.8 77.8 69.2 ground, these tiny spiders capture small soft- Aphididae 30.0 0.0 I l.l bodied insects, including numerous springtails Orthoptera 17.9 t2.2 0.r (Collembola), dipterans, and homopterans (Ta- Apis mellifera L. l.l 4.4 15.5 3). harmful cereal aphids can Other Apidae 2.1 0.0 0.0 ble Agriculturally Formicidae 8.4 0.0 0.0 form a signiffcant portion (-12-40Vo) in the prey Other Hymenoptera 1.0 0.0 2.2 of the dwarf spiders in European winter wheat Coleoptera 5.8 l.l 0.0 fields (Table 3) (Sunderland et al. 1986, Nyffeler Araneae 0.0 o.o 0.0 Benz 1988a). Green rice lea{hoppers,Nepho- Lepidoptera 0.0 0.0 0.0 & Others 6.9 1.2 1.9 tettix cincticeps (Uhler), and brown plant- Total r00.0 100.0 100.0 hoppers, Nilaparaata lugens (Stal), composed No. prey records r90 90 2t5 =60Vo of the prey of Oedothorar insecticeps Boes. & Str. in rice fields in Asia (Table 3; Kiri- " Argiope aurantia Lucas in cotton in Texas (Nyffeler et al. tani et al. 1972). 1987a). Mesh-Web Weaoers. Mesh-web weavers (Dic- b Argiope bruennichi (Scopoli) in grassland in Switzerland (Nyffeler 1982). tynidae) are small spiders (-3 mm in length) of 'A. bruennichi in gtassland in Switzerland (Nyffeler 1982). brownish, greyish, or green color that use the 1360 ENvInoxurNrer- ENrovror-ocv Vol. 23, no. 6

Table 3. Prey spectrum (in percent) of sheet-web weav- Table 4. Prey apectrum (in percenr) of comb-footed ere (Linyphiidae) baeed on three different field srudies spiders (Iheridiidae) baaed on three differenr field etudiee

Prey type Study l" Study 2b Study 3" Prey type Study l" Study 2b Study 3'

N ilap ans ata luge n s (Stal) 0.0 0.0 23.9 S o Ie nop si s intsic ta (Bur en) 0.0 /J.J 0.0 N e phot e ttix cincti c e p s (U hler) 0.0 0.0 38.9 Other Formicidae 92.0 0.4 0.0 Aphididae 38.7 lz.t Coleoptera 0.0 15.l 3.1 Collembola 37.8 7t.7 -d Diptera 0.0 0.0 27.8 Diptera r3.5 5.6 -d Aphididae 0.0 4.6 42.6 Thysanoptera _d 4.0 -d Cicadellidae 3.0 0.8 0.0 Araneae 0.0 1.5 -d16.3 Thysanoptera 0.0 0.0 9.0 0thers 10.0 5.1 20.9 Ephemeroptera 0.0 0.0 7.6 Total 100.0 100.0 100.0 Others 5.0 3.8 9.9 No. prey records lll 198 226 Total 100.0 100.0 100.0 No. prey records l0l 258 223 " Erigone spp.lOed.othorax spp. in winter wheat in Switzer- land (Nyffeler & Benz 1988a). " Achaearanea riparia (Blackwall) under overhanging grass b Various linyphiid species in winter wheat in England (Sun- in Switzerland (Nyffeler & Benz l98l). derland et al. 1986). b Latroilcctus nlactans (F.) in cotton in Texas (Nyffeler et al. ' Oedothorax insecticeps Boes. & Str. in rice ffelds in Japan r988). (Kiritani et al. 1972). " Theridion inpressum. L. Koch in wheat ffelds in Switzer- d Information not available. land (Nyffeler 1982). calamistrum to comb out cribellate si,lk from a Black widow spiders also frequently capture sieve-like plate just forward of the other spin- beetles (l1%o of total prey; Table 4) including the nerets called the cribellum. Soft-bodied insects, boll weevil, Anthonomus grandis grandis Bohe- predominantly small adult dipterans and ho- man (Whitcomb 1974, Ny$eler et al. 1988). The mopterans, are intercepted in the small irregular western widow spider, Latrodectus hesperus mesh webs that the dictynids spin on leaves of Chamberlin & Ivie, known to feed primarily on various ffeld crops and wild plants (Nyffeler & various beetles (Pratt & Hatch 1938), was ob- Benz 1981, Nentwig 1987). Agriculturally harm- served foraging on harvester ants in California ful dipterans and aphids can compose a high (MacKay 1982). Some theridiids (Steatoda spp. percentage in the diet of dictynid spiders with -6 mm adult length) feed primarily on a (Heidger & Nentwig f989). In other studies, dic- diet of various flies and meal-infesting insects in tynids were recorded foraging on small bugs stables and barns (Nyffeler & Benz 1987). Aphids (Heteroptera) (Nyffeler et al. 1992b). constituted -10-90Vo of the prey of theridiids in Comb-Footed Spiders. This family (Theridi- European field crops (Nyffeler & Benz 1981). idae) of small to medium-sized species, are char- Small kleptoparasitic theridiids, Arggrodes acterized by a globular abdomen. Theridiids spp. (-4 mm adult length), live in the webs of spin irregular webs and throw viscid silk on their other spider species and forage by stealing prey victim before biting it (Nentwig 1987). Theridi- from the host or taking prey below the threshold ids are, in general, exceedingly polyphagous ofacceptability (in size) ofthe host, or occasion- (Nyffeler & Benz 1981). However, in environ- ally attacking the host or its young (Nyffeler et al. ments where ants occur in large numbers, these 1987a, Vollrath 1987). spiders can switch to predominantly feeding on Funnel-Web Weaoers. These weavers (Age- ants (myrmecophagy; Table 4) (MacKay 1982, lenidae) trap their prey by means of funnel-like Nyffeler et al. 1988). Ants compose >907o of the sheet webs. At the entrance of the funnel, the prey of the European species Achaearanea ri- spider waits for prey. When an insect lands on paria (Blackwall) (-3.5 mm adult length) under the sheet, the spider runs quickly to the victim, overhanging grass (Table 4). Myrmecophagy was bites it, and carries it to the funnel entrance also observed in the southern black widow spi- where feeding takes place. Mass occurrences of der, Latrodectus rnactons (F.), a dangerously Agelena labgrinthica (Cl.), a dark brown Euro- venomous species whose black colored females pean species, with -10 mm adult length, can (=10 mm in length) show a distinct red hour- sometimes be seen in minimally disturbed grass- glass marking on the ventral part of the abdomen. land (old fields). In the strong extensive funnel L. mactans was observed to capture primarily webs, these spiders capture a wide variety of red imported ffre ants, Solenopsös önoöcta (B:u- different insect groups (Table 5), which includes ren), (75Vo oftotal prey; Table 4) in cotton ffelds at times numerous agriculturally harmful lepi- of East Texas, where this spider builds irregular dopterans from the family Pieridae. Honey bees, mesh type webs in holes in the ground, in large A. mellifera, and grasshoppers (Orthoptera) con- depressions between dirt clods on the ground stitute high proportions in the prey of this spider surface, or in the lowest branches of plants in some habitats (Table 5; Nyffeler 1982). (Nyffeler et al. 1988). Black widow immatures, In the litter of European woodlands, the dark third instar or older, can capture fire ant workers. brown Coelotes terrestris (Wider) (=I0 mm December 1994 Nyrrnr,sn ET AL.: How Spronns MAKE A LrvrNG I36l

Table 5, Prey spectrum (in percent) of funnel web- lynx spiders may capture an average of =l prey weaven (Agelenidae) based on three different field srudies per spider per day in the ffeld (Nyffeler & Benz f988b, Nyffeler 1992a). populations Prey type Study l' Study 2b Study 3' et al. Field of hunting spiders were observed in an underfed Apis mellifera L. 23.3 1.9 0.0 condition by researchers in North America, Eu- Fomicidae r3.3 D./ 1.0 Other Hymenoptera 8.3 1.9 1.0 rope, and Japan (see Nyffeler & Breene [f990al Orthoptera 0.0 26.4 0.0 for a review). Apparently, low feeding frequency Coleoptera 5.0 l.ö 64.0 is a pattern characteristic for spiders foraging Lepidoptera r8.3 37.7 0.0 (Zim- Diptera tt.7 15. r t7.o without a web in the natural environment Trichoptera 10.0 0.0 0.0 mermann & Spence 1989, Wise 1993). Signifi- Dermaptera 0.0 0.0 8.0 cantly higher feeding frequencies can be ob- Other 10. r 3.8 9.0 served in laboratory experiments when food is Total r00.0 t00.0 100.0 offered ad libitum (Nyffeler & Breene lgg0a, No. prey records 60 DJ 306 Nyffeler et al. 1992a). High levels of cannibal- " Agelena labgrinthica (Cl.) in grassland in Switzerland ism, observed in hunting spiders, may be crucial (Nyffeler 1982). for their survival under conditions of food limi- b A.labgrinthica in grassland in Switzerland (Nyffeler 1982). tation. 'Coelotes terrestris (Wider) in hedges in Gemany (Petto r990). Wolf Spiders. These spiders (Lycosidae) are small to large-sized animals, characterized by the specific arrangement of their eight eyes; they form three rows with the consisting adult length) builds tube-like funnels that end anterior row of four small several centimeters under ground. Most of the eyes and the two back rows consist- remains found in such ground funnels were the ing each of two larger eyes. These spiders are ground elytra of beetles (including numerous Cara- vagrant hunters that forage on the surface bidae), which indicates that C. temeströs concen- well-camouflaged by their brownish to greyish trates largely on beetle prey (Table 5; Nyffeler & coloration. Contrary to common belief, wolf spi- Benz 1981, Petto 1990). The hard-sclerotized ders do not necessarily run down their prey beetles are probably not optimal diet for most (Wise 1993). More recent studies suggest that smaller-sized spiders, because the chelicerae they tend towards a sit-and-wait foraging strat- cannot penetrate the thick cuticle of these in- egy. With their stout chelicerae they chew down sects (Nentwig 1987). Some spider species, their prey to a "meat ball" (Kiritani et al. 1972). which inhabit microhabitats rich in beetle prey Wolf spiders of the genus (5-8 mm adult such as C. terrestris, exhibit a specialized pred- length) are often characterized as diumal forag- atory behavior by biting into the intersegmental ers (e.g., Yeargan 1975); but nocturnal predation membranes of beetles (Nentwig 1987). Spiders activities could be monitored as well (Whitcomb that live in tubes under ground, Atypidae, 1974, Hayes & Lockley 1990). Pardosa spp. wolf Ctenizidae, and Eresidae (=10-15 ".g.,mm in length), spiders are abundant in field crops, grasslands, concentrate largely on beetle prey (Nyffeler & and woodlands where they forage on small soft- Benz l98l). bodied . Their diet includes spring- Another agelenid, the water spider Arggroneta tails (Collembola), small dipterans, and ho- aquatica Clerck (-10 mm adult length), Iives in mopterans (Table 6; Edgar 1970, Nyffeler & a bell (air bubble attached to water plant) under Benz 1988b, Nyfieler & Breene f990a). Agricul- water in ponds and streams. This palaearctic spe- turally harmful cereal aphids can constitute an cies mostly hunts fly larvae and small crusta- essential portion in the diet of Pardosa spp. in ceans (Foelix 1982). (Recently Arggroneta has European winter wheat ffelds (Table 6; Nyffeler been placed into its own family, Argyronetidae & Benz f988b). Leafhoppers and dipterans con- lPlatnick 1993]). stitute essential components in the diet of Par- Foraging Patterns of Hunters. Low feeding fre- dosa ramulosa (McCook) in ffeld crops in Cali- quency (-IUEI spiders feeding simultaneously fornia (Table 6; Yeargan 1975,Oraze & Grigarick in a given population) was observed in each of f989). In rice ffelds in Asia, green rice leaf- the four families of hunters described in this hoppers, N. cinctöceps, and brown planthoppers, article (wolf spiders, lynx spiders, crab spiders, N. lugens, composed -80Vo of the diet of wolf and jumping spiders) (Nyffeler & Breene 1990a). spiders (Table 6; Kiritani et al. 1972). Mosqui- With a visual method based on average percent- toes (Aedes), shore flies (Ephgdra), and bugs age of spiders with prey in their chelicerae ob- (waterboatman Trichocorixa) are the primary served in the ffeld, average hunting (searching) food source for P, ramulosa in marshes (Green- time, and handling time assessed in the labora- stone 1979). tory, the predation rate (number of prey per spi- Large nocturnal wolf spiders, genera Rabidosa der per day) ofa spider individual can be roughly and Hogna (previously known as Lgcosa, estimated (Nyffeler et al. 1987b). With this -15-20 mm adult length), often feed on bulky method we estimated that adult wolf spiders and prey including large grasshoppers, crickets, bee- 1362 ENvrnoNurNTAL ENToMoLocY Vol.23, no.6

Table 6. Prey spectrum (in percenr) of wolf spiders (Heteroptera) apparently are optimal diet for O. (Lycosidae) based on three different field studies salti,cus (Lockley & Young 1987; Agnew & Smith 1989; Breene et al. 1990; Nyffeler et al. 1992a, b). Prey type Study l" study 2b study 3" Lockley & Young (1987) reported that O. salticus N ilaparoat a lu ge ns (Stäl) 0.0 0.0 24.9 fed heavily on tarnished plant bugs, Lggus line- phot p (Uhler) N e ettir cinct ic e s 0.0 0.0 52.6 (P. de B.) (40% total prey), Other Cicadellidae 0.0 19.3 olaris of cotton flea- Aphididae 27.1 4.8 -d hoppers, Pseudatomoscelis seriatus (Reuter) Diptera 27.t 22.1 -d (8Vo), and other bugs (6Vo) in a cotton ffeld in Collembola 25.4 1.8 -d -d Mississippi (Table 7). In a cotton agroecosystem Heteroptera 0.0 r 1.5 preyed Orthoptera 0.0 6.3 -d in Texas, O. salticus heavily on cotton Coleoptera 5.1 6.0 -d fleahoppers (24Vo of total prey; Table 7), but in Araneae 6.8 19.6 -d8.9 another cotton area with low incidence of flea- Others 8.5 8.6 13.6 hoppers and other true bugs, red imported ffre Total 100.0 100.0 100.0 ants (S. öru:icta) were most frequently captured prey No. records 59 331 r,553 (22Vo oftotalprey; Table 7) (Nyffeleretal. 1987b, 1992a). " Pard,osa spp. in winter wheat in Switzerland (Nyffeler & Benz 1988b). Pollinating bees attracted to wild flowers and b Pardosa ramulosa (McCook) in alfalfa ffelds in California cotton plants during bloom are frequently en- (Yeargan 1975). countered and overpowered by the green lynx (=LVcosa) " Pardosa pseudoannulata (Boes. & Str.) in rice spider Peucetia oiridans (Hentz), a larger sized ffelds in Japan (Kiritani et al. 1972). d Information not available. aggressive species (up to >15-mm length), that lies in ambush on the upper surface of leaves well camouflaged by its bright gleen color and tles, noctuid moths, and other spiders (Van Hook cryptic posture. Bees (including A. mellifura) 1971, Whitcomb 1974, Hayes & Lockley 1990). constituted 23Vo ofthe prey ofgreen lynx spiders (Oxyopidae) Lgnx Spöders. These predators in a Texas cotton ffeld; theie spiders also piey on are characterized by the erect long spines on pests such as cotton fleahopper and boll weevil their legs and by a hexagonal eye arrangement. (Nyffeler et al. 1992a). Lynx spiders can be active day or night (Nyffeler Crab Spiders. These spiders (Thomisidae) are et al. 1987b). The striped lynx spider, Oxgopes a family of small to medium-sized species of spi- salticus lJentz, a light-colored species with an ders characterized.by their crab-like posture and average adult length of :6 mm, was found to be walking behavior (like crabs they walk laterally). the most abundant spider predator in cotton Crab spiders are among the most abundant spi- fields and other agricultural crops in parts ofthe der predators in grasslands and agricultural southern United States (Dean & Sterling 1987, crops. They are considered to be typical sit-and- Young & Edwards 1990). O. saltöcus is a pounc- wait foragers that lie motionless in ambush for ing hunter that actively searches the plant sur- prey. McDaniel & Sterling (1982), however, pro- face for prey. This spider captures a wide variety vided evidence that crab spiders may at times of small-sized arthropods (up to -6 mm maxi- actively search for prey (feeding on immobile mum prey length) and shows considerable flexi- insect eggs). Feeding can take place day or night. bility in switching its dietary composition in re- Brown colored crab spiders of the genus Xysti- sponse to prey availability (Table 7). Small bugs cus (-f mm adult length) feed on small winged Hymenoptera and Diptera most frequently when Table 7, Prey apectrum (in percent) of lynx spider, observed on meadow plants (Table 8); those on Oxyopes salticus Hentz (Oxyopidae), based on three dif- the soil surface prey more often on ants, spiders, ferent field atudiea carabid beetles, and springtails (Table 8; Nyffeler & Breene f990b). Early-instar crab spi- Prey type Study l" Study 2b Study 3' ders feed on soft-bodied insects such as tiny P s eudatomo sc eli s I e riatus dipterans, hymenopterans, thrips, (Reuter) aphids, and 8.3 23.8 0.0 whereas Lggus lineolaris (P. de B.) 39.6 1.6 0.0 later instars and adults occasionally Other Heteroptera 6.2 9.5 4.7 overpower large and well-armed insects includ- Diptera 18.7 r5.9 t7.2 ing large stinging bees. Large bees comprised Aphididae 0.0 t2.7 l4.t <1Vo of the total prey of Xgsticus spp. in hay Cicadellidae t6.7 0.0 r7.2 meadows (Nyffeler & Breene f990b). Morse's S ole no p s is inaic ta (Bw en) 0.0 9.5 21.9 Lepidoptera 6.2 0.0 0.0 (1983) quantitative prey analysis listed large Araneae 0.0 15.9 l4.l bees (A. mellifera, Bombus spp.) as comprising Others 4.3 ll.l 10.8 =\OVo (by numbers) of the natural diet of Misum- Total 100.0 100.0 100.0 ena oatia (Clerck) (Table 8). This white, yellow, No. prey records 48 6364 or pale green colored spider of -10 mm adult length is perfectly camouflaged on flowers where " In cotton in Mississippi (Lockley & Young 1987). b In cotton in Central Texas (Nyffeler et al. 1992a). it waits in ambush for pollinating insects. Mis- 'In cotton in East Texas (Nyffeler et al. 1987b). unxenops celer (Hentz) (-6 mm adult length) December 1994 Nyrppr.nn ET AL.: How Sprorns MAKE A LIVTNG 1363

Table 8. Prey spectrum (in percenr) of crab spiden to other prey groups. form facultative (Ihomiaidae) A of based on three difrerent field srudiee monophagy was observed in the jumping spider Plexöppus pagkulli (Audouin), cosmopolitan Prey type Study l" Study 2b Study 3" a species of :10 mm adult length, that takes up Diptera il.8 0.0 7.O residence in and on buildings and rarely mi- Apidae 4.O 0.0 49.3 Formicidae 4.8 34.3 _d grates into field crops. This species is known Other Hymenoptera 8.0 oo 4.2 from the literature as a polyphagous feeder on Lepidoptera 1.6 0.0 29.6 a wide variety of arthropod taxa including Coleoptera 4.8 8.6 _d Odonata, Orthoptera, Homoptera, Lepidoptera, Aphididae 0.0 I r.4 _d Collembola 0.0 J.t _d Diptera, Hymenoptera, and other Araneae (Jack- Araneae 6.4 25.7 _d son & MacNab f989, Nyffeler et al. 1990b). How- Others 5.6 I1.4 9.9 ever, in a roach-infested building in Central Total 100.0 100.0 r00.0 Texas this spider was observed to concentrate No. prey records 125 35 7t largely on the German cockroach, Blattella ger- manica (L.), as a food source (>9OVo of total prey; " Xgsticus spp. on meadow plants in Switzerland (Nyffeler 1982). Table 9); regardless of the highly limited diet, b Xysticus spp. on soil surface of meadows in Switzerland the P. pagkulli females produced viable of- (Nyffeler 1982). spring, which implies that c-Misumena the nutritional quality uatia (Clerck) on flowers in Maine (Morse lg83). of the food supply was sumcient for the spiders' d Information not available. growth and reproductive needs (Nyffeler et al. r9gob). jumping feeds readily on various bugs in field crops Another member of the spider family, (Dean et al. 1987, Agnew & Smith 1989, Breene Phidippus audax (Ilentz) (=10 mm adult length), et al. 1990). Crab spiders were reported feeding is one of the most abundant spider predators in (Young on the Colorado potato beetle (Cappaert et al. ffeld crops in the United States & Ed- 19er). wards 1990). P. audax feeds heavily on agricul- Jumping Spiders. This is a family (Salticidae) turally harmful bugs such as cotton fleahoppers of small to large-sized species with rectangular and tarnished plant bugs (Table 9; Dean et al. shape, short stout legs, and greatly enlarged an- 1987, Young f989). This spider demonstrated a terior median eyes. Possessing acute vision these sigmoid functional response to the availability of diurnal hunters react to visual stimuli such as fleahopper prey in field conffnement tests (Breene passing insects (Foelix 1982). They crawl to et al. 1990). P. audax also preys on bee- within striking distance and then jump on their tles (e.g., spotted cucumber beetle and boll wee- prey with great accuracy. Spiders of this family vil) and larvae of the bollworm, Helicooerpa zea (Boddie) (Young are highly polyphagous (Table 9) but can narrow 1989). Jumping spiders fre- their prey spectrum significantly, when a suit- quently eat other spiders (Jackson f977) (Table able prey species reaches high numbers relative 9). In different parts of the world, jumping spi- ders were observed feeding on insect eggs (Whitcomb 1974, McDaniel & Sterling 1982, Table 9. Prey epectrum (in percent) of jumping spi- Nyffeler et al. 1990a). Some salticid species (Por- den (Salricidae) baeed on three difrerent field etudies tia spp.) habitually invade the webs of other spi- ders and eat the web owners (araneophagy) Prey type Study l" Study 2b Study 3" fiackson & Blest 1982). Members of the family Blattella germanica (L.) 0.0 0.0 96.4 Mimetidae (pirate spiders) are known to prey P s eud,at omo s c elis s e riatu s exclusively on other spiders in the ffeld (Foelix (Reuter) 44.4 0.0 0.0 Lsgus lineolaris (P. de B.) 0.0 22.4 0.0 1982, Agnew & Smith 1989, Wise 1993) but in Other Heteroptera 2.8 8.6 0.0 the laboratory some mimetids feed on insects as Diptera 2.8 L2.l 0.0 well (Nentwig f987). Cicadellidae b.b 0.0 0.0 The feeding behaviors of other spider groups Membracidae 0.0 15.5 0.0 Hymenoptera 8.3 r.7 0.0 are discussed elsewhere (e.g., Nentwig 1987, Lepidoptera 8.3 1.7 0.0 Wise 1993). The wide variety of spider diets Coleoptera 0.0 22.4 0.0 shown in Tables l-9 reflects the diversity and Orthoptera b.t) 0.0 3.6 fexibility of foraging behavioral patterns utilized Araneae ooo 15.5 0.0 Others 0.0 0.0 0.0 by these animals in their quest for food. Total 100.0 100.0 100.0 Ecological Implications of the Insectivorous No. prey records 36 58 28 Activities of Spiders. As generalist predators, spi- ders destroy pest insects, insects of a neutral " Phidippus audnx (Hentz) on wild plants and cotton in economic status, and beneffcials alike (Bilsing Texas (Dean et al. 1987 and M.N. unpublished data). 1920, Whitcomb 1974, Nyffeler 1982). The same b P. auilar on wild plants and cotton in Mississippi (Young 1989). spider species that feeds predominantly on pests c Plerippus pagkulli (Audouin) in building in Texas at a certain location, may capture mostly benefi- (Nyffeler et al. 1990b). cials at another location only a few kilometers 1364 ENvlnoNueurAl ENToMoLocY Vol.23, no.6 away. The orb weaver Argiope bruennichi (Sco- Table 10. Prey kill of apiders in various ecosyalema instance preys primarily on acridid (rcugh eetimates computed from literature dara [modified poli) for after Nyffeler 19821) grasshoppers in some grasslands in Central Eu- rope; however, in small old fields dominated by Prev kill L;eosraDhlc. area , ; flowering thistles and blackberry bushes, this Ecosystem Kgtnuyf spider was observed capturing large numbers of Field crops and mown honey bees (Nyffeler & Benz I98l). Honey bees, meadows" Central Europe =2 bumble bees, and other pollinating insects are a Phragmites reed belt of primary food source for some aggressive spiders lake (mown once/year)ä Central Europe -5-f0 that search and wait for prey on or near flowering Minimally disturbed plants (see above) (Bilsing 1920, Nentwig 1987, grassland (old ffeld)' United States -50 Nyffeler & Breene 1991). Agnew & Smith (1989) Minimally disturbed and Nyffeler et al. (1987b, 1992a) observed that grassland (old ffelds) and forestsd Central Europe in ffeld crops in the southwestern United States, -l0o-f50 spiders frequently kill and eat other predators Marsh land' United States -200 (intraguild predation). Whitcomb (1974) stated Tropical coffee plantation Oceanic-Australian that some web weaving spiders destroy large (insecticide-freeI region -160 numbers of parasitoids and predators. These For purposes of comparison all estimates are converted to negative e$ects, however, are balanced by spi- Kilograms (freshweight)/ha/yr. Assumptions: =807o of the der activities in killing numerous pest insects as killed prey is consumed; caloric equivalent of prey is -5.6 (Moulder well (for a discussion see Agnew & Smith If989], cal/mg dry weight & Reichle 1972). " Kajak et al. (1971); Luczak (1975). Nyffeler & Breene [1991], and Nyffeler et al. b Prihringer (1979). [f987b, 1992a, b]). Furthermore, predation on 'Van Hook (197I). beneffcials may be helpful in maintaining the d Kaiak.t al. (197r); Stern & Kullmann (1975). number of spiders during a period of food short- 'Teal (1962). fRobinson & Robinson (1974). age (low pest levels). Although the ecological significance of spiders in the balance of nature is still largely unex- Dean & Sterling 1987). These methods do not plored, they generally are considered to be im- take into account those spiders that inhabit portant natural enemies of insects (Robinson & cracks in the soil between the rows, and the Robinson 1974, Whitcomb 1974, Zimmermann & available data from ffeld crops are, therefore, Spence 1989, Young & Edwards 1990). Turnbull rather conservative estimates. Nevertheless, (1973) surveyed 37 published censuses ofspider mean spider densities in U.S. crops are signiff- numbers in a wide variety of natural and modi- cantly lower than Turnbull's overall mean value ffed environments. He found an overall mean of 130.8/m2 (see above). Field crops are highly density of 130.8 spiders per square meter (range, disturbed systems whose beneffcial aühropod O.6-8421m2) and concluded that spiders must numbers are drastically reduced by agricultural have an enorrnous predation impact on insect practices such as frequent mowing, cultivating, populations. Especially in minimally disturbed combine-harvesting, and use of heavy doses of systems such as old fields, marshes, and wood- pesticides (Luczak 1979, Nyffeler 1982, Riechert lands colonized by spiders all year long in & Lockley 1984). high numbers (up to a maximum of -1,000/m2) In the literature, methods by which predator (see Dondale 1971), these animals seem to play numbers in an agroecosystem could be increased an important ecological role as insectivores are discussed (Nyffeler 1982, Sterling et al. 1989, (Nyffeler & Benz f987). The prey kill by the Wise 1993). Young & Edwards (1990) suggest spiders of such ecosystems was estimated at several management strategies (e.g., reduction of =50-200 kg fresh weight per hectare per year pesticide usage and cultivation frequencies) that (Teal 1962, Kajak et al. 1971, Van Hook 1971, could enhance the spider numbers in ffeld crops Stern & Kullmann 1975), which may be -100 and adjacent habitats resulting in increased pre- times higher compared with average agricultural dation activities. In Japan, attempts have been fields of the temperate-northern zones (Kajak et made to raise the fecundity of spiders in rice al. f97f, Luczak 1975, Nyffeler 1982) (Table l0). ffelds artiffcially by releasing Drosophila flies as Nyffeler et al. (1994) surveyed 25 censuses of a supplementary food supply; this caused an in- spider numbers in U.S. ffeld crops published by crease in spider numbers (i.e., augmentation of ll different research groups (considering a geo- natural enemies) (Kobayashi 1975). graphic range from North Carolina to California), There is evidence that spiders may play an which gave an overall mean density of -1 plant- important role as mortality agents of certain crop dwelling spider per square meter (t0.18 SEM). pests of small body size such as aphids (Aphid- Spider numbers in cotton throughout Texas av- idae), leafhoppers (Cicadellidae), planthoppers eraged 0.8/m2 (Dean & Sterling 1987). Such es- (Delphacidae), and fleahoppers (Miridae) in timates are based on D-Vac samples, whole plant some agricultural fields where little or no insec- sampling, and ground cloth technique (e.g., ticide is used (Kiritani et al. 1972, Liao et al. December 1994 Nvprrr-rn ET AL.: How Sprosns MAKE A LrvrNG 1365

1984, Oraze & Grigarick 1989, Nyffeler et al. fleahopper [Hemiptera: Miridae]. Entomophaga 35: f992b). Robinson & Robinson (1974) estimated 393-40r. that spiders may destroy the equivalent of -160 Cappaert, D. L., F. A. Drummond & P, A. Logan. kg insects per hectare per year in an insecticide- 1991. Population dynamics of the Colorado potato free coffee plantation in New Guinea (Table t0). beetle (Coleoptera: Chrysomelidae) on a native These authors cautiously conclude that the ab- host in Mexico. Environ. Entomol. 20: f549-1555. Coddington, A. W. Systematics sence of coffee pests in their study area may be, J. & H. Levi. 1991. and evolution of spiders (Araneae). Annu. Rev. at least in part, attributable to the collective pre- Ecol. Syst. 22: 565-592. dation impact of the rich spider fauna. Sterling et Culin, D. & K. V. Yeargan. 1982. Feeding behav- (1992) J. al. demonstrated with computer model- ior and prey of Neoscona arabesca (Araneae: Ara- ling techniques that the insectivorous activities neidae) and Tetragnatha laboriosa (Araneae: Tet- of spiders and other arthropod predators are of ragnathidae) in soybean ffelds. Entomophaga 27: economic value in certain years in unsprayed 4r7-424. cotton in Texas. Experimental evidence for the Dean, D. A. & W. L. Sterling. f985. Size and phe- ecological impact of spiders has been reviewed nology of ballooning spiders at two locations in in detail by Wise (1993) (see his book for original eastern Texas. J. Arachnol. 13: lll-I20. citations). 1987. Distribution and abundance patterns of spi- Coddington & Levi (1991) state that the order ders inhabiting cotton in Texas. Tex. Agric. Exp. Araneae ranks seventh in global diversity after stn. Bull. 8-1566. Dean, D. A., W' Sterling, Nyffeler R. G. the five largest insect orders (Coleoptera, Hyme- L. M. & Breene. 1987. Foraging by selected spider pred- noptera, Lepidoptera, Diptera, Hemiptera), and ators on the cotton fleahopper and other prey. the order Acari. Wise (1993) considers Southwest. Entomol. 12: 263-270. 'model the spider a predator' in terrestrial eco- Dondale, C. D, 1971. Spiders of Heasman's ffeld, a systems. Van Hook (1971) and other ecologists mown meadow near Belleville, Ontario. Proc. En- recognized that spiders as secondary consumers tomol. Soc. Ont. l0l: 62-69. "may contribute significantly in maintaining Eberhard, W. G, 1990. Function and phylogeny of community homeostasis." Spiders play an inte- spider webs. Annu. Rev. Ecol. Syst. 2L:341-372. gral part in herbivore- and detritus-based food Edgar, W' D. f970. Prey and feeding behaviour of chains in terms of biomass, energy flow, and nu- adult females of the Pardosa ament&ta trient transfers (Turnbull 1973, Schoener 1989, (Clerck). Neth. J. Zoo1.2O: 487-491. Wise 1993). Surprisingly, the basics of spider Eisner, T., R. Alsop & P. Ettershank. 1964. Adhe- siveness spider Science (Washington, predation ecology (i.e., prey preferences, search of silk. DC) 146: 1058-1061. areas, search times, handling times, predation Foelix, R. F. 1982. Biology of spiders. Harvard Uni- rates, functional and numerical responses) are versity Press, Cambridge, MA. still largely unknown for most species. Further Greenstone, M. H. 1979. Spider feeding behaviour detailed investigations on the predatory role and optimises dietary essential amino acid composition. economic impact of spiders in various natural Nature (Lond.) 282: 501-503. and agricultural habitats are urgently needed. Greenstone, M. H. & A. F. Bennett. f980. Foraging With this article we hope to generate some inter- strategy and metabolic rate in spiders. Ecology 6l: est among entomologists and ecologists for fu- 1255-1259. fure studies on spider impact. Hayes, J. L. & T. C. Lockley. 1990. Prey and noc- turnal activity of wolf spiders (Araneae: Lycosidae) in cotton ffelds in the Delta region of Mississippi. Acknowledgments Environ. Entomol. 19: l5l2-I518. Heidger, C. & W. Nentwig. 1989. Augmentation of We thank Marvin K. Harris and Willard H. Whit- beneffcial arthropods by strip-management. 3. Arti- comb for manuscript review. Two anonymous review- ffcial introduction of a spider species which preys ers offered valuable criticisms. This work was partially on wheat pest insects. Entomophaga 34r 5IL-522. supported by a postdoctoral fellowship of the Swiss Jackson,R.R. L977. PreyofthejumpingspiderPhi- National Science Foundation awarded to M.N. and dippus johnsoni (Araneae: Salticidae). f. Arachnol. by Expanded Research Projects H-2591-2I00 and 5: 145-149. H-6903-2100 of the Texas Agricultural Experiment Jackson, R. R. & A. D. Blest. 1982. The biology of Station. Approved for publication as TA 31513 by Di- Portia a web-building jumping spider rector, Texas Agricultural Experiment Station. fimbriatla, (Araneae, Salticidae) from Queensland: utilization of webs and predatory versatility. l. Zool. (Lond.) References Cited 196: 255-293. Jackson, R. R. & A. MacNab. 1989. Display and Agnew, C. W. & J. W. Smith, Jr. f989. Ecology of predatory behaviour of Plexippus pagkulli, a jump- spiders (Araneae) in a peanut agroecosystem. Envi- ing spider (Araneae, Salticidae) from Florida. N.Z.J. ron. Entomol. 18: 30-42. Zool. 16: 15l-168. Bilsing, S. W. 1920. Quantitative studies in the food Kajak, 4., A, Breymeyer & J. Petal. f971. Produc- ofspiders. Ohio J. Sci.20: 2L5-26O. tivity investigation of two types of meadows in the Breene, R. G., W. L. Sterling & M. Nyffeler. 1990. Vistula Valley. Predatory arthropods. Ekol. Pol. A Efficacy of spider and ant predators on the cotton 19:223-233. 1366 ENvrnoNNasNTAL ENToMoLocY Vol. 23, no. 6

Kiritani, K., S. Kawahara, T. Sasaba & F, Nakasuji. 1990b. Spiders associated with selected European 1972. Quantitative evaluation ofpredation by spi- hay meadows and the effects ofhabitat disturbance, ders on the green rice leafhopper,Nephotettir cinc- with the predation ecology ofthe crab spiders, Xgs- ticeps lJhler, by a sight-count method. Res. Popul. ticus spp. (Araneae: Thomisidae). J. Appl. Entomol. Ecol. 13: 187-200. 110:149-159. Kobayashi, S, 1975. The effect of Drosophila re- 1991. Impact of predation upon honey bees (Hyme- lease on the spider population in a paddy ffeld. noptera, Apidae), by orb-weaving spiders (Araneae, Appl. Entomol. Zool. IO : 268 -27 4. Araneidae and Tetragnathidae) in grassland ecosys- LeSar, C. D. & J. D. Unzicker. 1978. Life history, tems. J. Appl. Entomol. lll: 179-189. habits, and prey preferences of Tetragnatha labori- Nyffeler, M., D. A. Dean & W. L. Sterling. 1987a. osa (Araneae: Tetragnathidae). Environ. Entomol. Feeding ecology ofthe orb-weaving spider Argiope 7: 879-884. aurantia (Araneae: Araneidae) in a cotton agroeco- Liao, H.-T., M. K. Harris, F. E, Gilshap, D. A. Dean, system. Entomophaga 32: 367 -37 5. C. W. Agnew, G. J. Michels & F. Mansour. 1984. f 987b, Evaluation of the importance of the striped Natural enemies and other factors affecting seasonal lynx spider, Oxgopes salticus (Araneae: Oxy- abundance of the blackmargined aphid on pecan. opidae), as a predator in Texas cotton. Environ. En- Southwest. Entomol. 9: 404-420. tomol. 16: ll14-1123. Lockley, T. C. & O. P. Young. 1987. Prey of the 1988. The southern black widow spid,et, Latrodec- striped lynx spider, Oxgopes salticus (Araneae, Ox- tus mactans (Araneae: Theridiidae), as a predator of yopidae), on cotton in the Delta area of Mississippi. the red imported ffre ant, Solenopsis inrsicta (Hy- J. Arachnol. 14: 395-397. menoptera:O Formicidae), in Texas cotton ffelds. J. Luczak, J. 1975. Spider communities of the crop- Appl. Entomol. LO6: 52--57. ffelds. Pol. Ecol. Stud. l: 93-110. 1989. Prey selection and predatory importance of 1979. Spiders in agrocoenoses. Pol. Ecol. Stud. 5: orb-weaving spiders (Araneae: Araneidae, Ul- r5r-200. oboridae) in Texas cotton. Environ. Entomol. 18: MacKay, W. P. 1982. The effect of predation of 373-380. western widow spiders (Araneae: Theridiidae) on Nyffeler, M., R. G. Breene, D. A. Dean & W. L. Ster- harvester ants (Hymenoptera: Formicidae). Oecolo- ling. 1990a. Spiders as predators of arthropod gia (Berl.) 53: 406-411. eggs. I. Appl. Entomol. 109: 490-501. McCormick, S. & G. A. Polis. 1982. Arthropods that Nyffeler, M., R. G, Breene & D, A. Dean. 1990b. prey on vertebrates. Biol. Rev. 57: 29J8. Facultative monophagy in the jumping spid,er, Pler- McDaniel, S. G. & W. L. Sterling. 1982. Predation ippus paykullt (Audouin) (Araneae: Salticidae). of Heliothis airescens (F.) eggs on cotton in east Peckhamia 2:92-96. Texas. Environ. Entomol. ll: 60-66. Nyffeler, M., D. A. Dean & W. L. Sterling. 1992a. Morse, D. H. 1983. Foraging patterns and time bud- Diets, feeding specialization, and predatory role of gets of the crab spiders Xgsticus emertoni Keyser- two lynx spiders, Orgopes salticus and Peucetia ling and Misumena aatia (Clerck) (Araneae, Tho- oiridans (Araneae: Oxyopidae), in a Texas cotton misidae) on flowers. J. Arachnol. 1l: 87-94. agroecosystem. Environ. Entomol. 2l: 1457-1465. Moulder, B. C. & D. E. Reichle, 1972. Signiffcance Nyffeler, M., W.L. Sterling & D. A. Dean. 1992b. of spider predation in the energy dynamics of Impact of the striped lynx spider (Araneae: Oxy- forest-floor arthropod communities. Ecol. Monogr. opidae) and other natural enemies on the cotton 42:473-498. fleahopper (Hemiptera: Miridae) in Texas cotton. Nentwig, W' 1987. The prey of spiders, pp. 249- Environ. Entomol. 21: l178-1188. 263. InW. Nentwig [ed.], Ecophysiology of spiders. 1994. Insectivorous activities ofspiders in U.S. ffeld Springer, Berlin. crops. f. Appl. Entomol. (in press). Nyffeler, M. 1982. Field studies on the ecological Oraze, M. J. & A. A. Grigarick. f 989. Biological con- role of the spiders as predators of insects in agro- trol of aster lea{hopper (Homoptera: Cicadellidae) ecosystems. Ph.D. dissertation; Swiss Federal In- and midges (Diptera: Chironomidae) by Pardosa stitute of Technology (ETH) , Zurich. ramulosa (Araneae: Lycosidae) in California rice Nyffeler, M. & G. Benz. 1981. Field studies on the ffelds. J. Econ. Entomol. 82 745-7 49. feeding ecology of spiders: observations in the re- Petto, R. 1990. Abundance and prey of Coelotes ter- gion of Zurich (Switzerland). Anz. Schaedlingskd. resfris {Wider) (Araneae, Agelenidae) in hedges. Pflanzenschutz Umweltschutz 54: 33-39 (in Ger- Bull. Br. Arachnol. Soc.8: 185-193. man). Platnick, N, I. 1993. Advances in spider f987. Spiders in natural pest control: a review. j. 1988-199f, with synonymies and transfers 1940- Appl. Entomol. 103: 321-339. 1980. New York Entomological Society, New York. 1988a. Prey and predatory importance of micry- Pratt, R. Y. & M. H. Hatch. 1938. The food of the phantid spiders in winter wheat fields and hay black widow spider on Whidby Island, Washington. meadows. J. Appl. Entomol. 105: 190-197. J.N.Y. Entomol. Soc. 46: l9f-193. f988b, Feeding ecology and predatory importance Pühringer, G. f979. Produktionsbiologie der Schilf- df wolf spiders (Pardosa spp.) (Araneae, Lycosidae) spinnen des Neusiedler Sees. Zool. Jahrb. Syst. in winter wheat ffelds. I. Appl. Entomol. 106: 123- 106: 500-528. 134. Riechert, S. E. & T. Lockley. 1984. Spiders as bio- Nyfreler, M. & R. G. Breene. 1990a. Evidence of logical control agents. Annu. Rev. Entomol. 29: low daily food consumption by wolf spiders in 299-320. meadowland and comparison with other cursorial Robinson, M. H. & B. Robinson. 1974. A census of hunters. I. Appl. Entomol. ll0: 73-81. web-building spiders in a coffee plantation at Wau, December 1994 Nvrrulrn ET AL,: How Sproens MAKE A LTvING 1367

New Guinea, and an assessment of their insecti- pp. 123-159. In P. N. Witt & J. S. Rovner [eds.], cidal effect. Trop. Ecol. 15: 9L107. Spider communication. Mechanisms and ecological Rovner, J, S. 1975. Sound production by Nearctic signiffcance. Princeton University Press, Princeton, wolf spiders: a substratum-coupled stridulatory NJ. mechanism. Science (Washington, DC) 190: 1309- Van Hook, R. I. l97f' Energy and nutrient dynam- 1310. ics of spider and orthopteran populations in a grass- Rovner, J. S. & F, G. Barth. 1981. Vibratory commu- land ecosystem. Ecol. Monogr. 4I I-26. nication through living plants by a tropical wander- Vollrath, F. 1987. Kleptobiosis in spiders, pp.274- ing spider. Science (Washington, DC) 2I4: 464- 286. In W. Nentwig Ied.], Ecophysiology of spiders. 466. Springer, Berlin. Schoener, T. W. 1989. Food webs from the small to Whitcomb, W. H. 1974. Natural populations of en- the large. Ecology 70: 1559-1589. tomophagous arthropods and their effect on the Sterling, W. L., K. M. El-Zik & L. T. Wilson. f989. agroecosystem, pp. 150-169. In F' G. Maxwell & F. Biological control ofpest populations, pp. 155-189. A. Harris [eds.], Proceedings, Summer Institute on In R. Frisbie, K. E]-Zik & L. T. Wilson [eds.], Inte- Biological Control of Plant Insects and Diseases. grated pest management systems and cotton pro- University Press of Mississippi, fackson, MS' duction. Wiley, New York. Wise, D, H. 1993. Spiders in ecological webs. Cam- Sterling, W. L., A. Dean & N. M. Abd El-Salam. f 992. bridge University Press, Cambridge, UK. Economic beneffts of spider (Araneae) and insect Yeargan, K. V. 1975. Prey and periodicity of Par- (Hemiptera: Miridae) predators of cotton fleahop- dosa ramulosa (McCook) in alfalfa. Environ. Ento- pers. j. Econ. Entomol. SS:52-57. mol. 4: 137-141. Stern, H. & E. Kullmann. 1975. Leben am seidenen Young, O. P. 1989. Field observations of predation Faden. Bertelsmann, Munich. by Phidippus audax (Araneae: Salticidae) on arthro- Sunderland, K. D., A. M. Fraser & A.F.G. Dixon. pods associated with cotton. J. Entomol. Sci. 24: 1986. Distribution of linyphiid spiders in relation 266-273. to capture ofprey in cereal ffelds. Pedobiologia 29: Young, O. P. & G. B. Edwards' 1990. Spiders in 367--375. United States field crops and their potential effect Teal, J. M. 1962. Energy fow in the salt marsh eco- on crop pests. j. Arachnol. 18: L-27. system of Georgia. Ecology 43: 6L4-624. Zimmermann, M. & J. R. Spence. 1989. Prey use of Tumbull, A. L. 1960. The prey of the spider the ffshing spider Dolomedes triton (Pisauridae, Lingphia triangularis (Clerck) (Araneae, Linyphi- Araneae): an important predator of the neuston idae). Can. I. Zool. 38: 859-873. community. Oecologia (Berl.) 80: 187-194. 1973. Ecology ofthe true spiders (). Annu. Rev. Entomol. l8: 305-348. Uetz, G. W, & G. E. Stratton. 1982. Acoustic com- Receiaed for publication 18 April 7994; accepted 28 munication and reproductive isolation in spiders, Juls 1994.