Biological Control 67 (2013) 390–396

Contents lists available at ScienceDirect

Biological Control

journal homepage: www.elsevier .com/ locate/ybcon

A potential spider natural enemy against virus vector leafhoppersin agricultural mosaic landscapes – Corroborating ecologicaland behavioral evidence

1049-9644/$ - see front matter � 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.biocontrol.2013.08.016

⇑ Corresponding author.E-mail address: samu.ferenc@agrar.mta.hu (F. Samu).

Ferenc Samu a,⇑, Orsolya Beleznai a,b, Gergely Tholt a,c

a Plant Protection Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, 15 Herman Ottó Str., Budapest, H-1022 Hungaryb Doctoral School of Animal- and Agricultural Environmental Sciences, University of Pannonia, Georgikon Faculty. 16 Deák Ferenc Srt., Keszthely, H-8360 HungarycEötvös Loránd University, Faculty of Scinece, Institute of Biology, Department of Systematic Zoology and Ecology. 1/C Pázmány Péter sétány, Budapest, H-1117 Hungary

h i g h l i g h t s

� Virus vector leafhopper moves fromcereal fields to grassy habitats atharvest.� A dominant hunting spider in grassy

margins has peak population densityat that time.� In laboratory the leafhopper is among

the favorable prey types for the spider.� When fed on a pure leafhopper diet

spiders maintained their growth.� The spider is a potential natural enemy

against the leafhopper.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 5 March 2013Accepted 29 August 2013Available online 5 September 2013

Keywords:SpiderLeafhopperVirus vectorNatural enemyPredationField marginPrey preference

a b s t r a c t

We intended to establish the potential for interaction between the wheat dwarf virus (WDV) vector leaf-hopper Psammotettix alienus – a dominant sap feeding pest in cereal fields, and the spider Tibellus oblon-gus – a dominant predator of grassy field margins. The relationship is important, because with thesenescing and harvest of cereals P. alienus migrates to alternative host species, grasses. We analyzedthe potential of T. oblongus to be an effective natural enemy of P. alienus by studying the probability oftheir co-occurrence seasonally and at the habitat and microhabitat scale. By gathering data from longterm research (1994–2011) in six agricultural regions of Hungary, we assembled 96 one-year-long data-sets obtained by suction sampling from the four key habitats of the agricultural landscape mosaic. Theanalysis showed that both in space and time the spider has the potential to prey on P. alienus. T. oblonguspopulations can reach considerable densities and represent high dominance among other spiders in thehabitats of the leafhopper. Given this co-occurrence pattern, we devised laboratory experiments to studywhether P. alienus is included among the preferred prey types of T. oblongus and to ascertain whether pro-longed feeding has no adverse effects and provides the nutrients for growth. P. alienus proved to be both apreferred prey type and one that can be utilized for growth by the spider. This study collected the circum-stantial ecological and direct laboratory feeding trial proofs that T. oblongus can be an important biolog-ical control agent against the leafhopper pest P. alienus.

� 2013 Elsevier Inc. All rights reserved.

1. Introduction

Recent field studies drew our attention to a leafhopper and aspider species, which are dominant herbivorous and predatory

F. Samu et al. / Biological Control 67 (2013) 390–396 391

arthropods in their respective habitats, cereal fields and field mar-gins. Leafhoppers have to seek alternative habitats around harvestand ploughing, among them grassy field margins. Therefore, weaimed to study the feasibility that the spider Tibellus oblongus(Walckenaer, 1802) (Araneae, Philodromidae) has the potential toact as a natural enemy against the leafhopper Psammotettix alienus(Dahlbom, 1850) (Auchenorrhyncha, Cicadellidae).

Predatory arthropods, like spiders, play an important role inregulating population dynamics of insects (Nyffeler and Sunder-land, 2003; Riechert and Lockley, 1984; Rypstra and Marshall,2005). For predation to take place predators and prey must occurtogether in space and time. Movements between different habitatsin landscapes and the seasonality of the interacting species influ-ences predation success. Agricultural fields have a typical spatialand temporal disturbance pattern. Regular management of cropfields, like harvest and ploughing, causes repeated changes in veg-etation architecture, cover and biomass. The ‘cyclic colonization inpredictably ephemeral habitats’ hypothesis (Wissinger, 1997) pre-dicts that, on the one hand, successful natural enemies must beadapted to this cyclic nature of agricultural habitats (Welchet al., 2011). On the other hand, pest species face the same recur-rences of disturbance, and are also likely to have evolved adaptivestrategies to cope with cyclic disturbance. Ripening of seeds andsenescing in crop plants negatively influences sap feeders (Hubertyand Denno, 2004; Williams, 1995), and might be the driving forcefor host alteration (Sandstrom, 2000), which culminates in the ma-jor disturbance of harvest and subsequent ploughing that forcesherbivorous insects to migrate to alternate or secondary habitats(Christian and Willis, 1993; Kindler et al., 1999).

In Central European cereal fields one of the most common spe-cies of leafhoppers is P. alienus. This leafhopper species is an impor-tant pest not for the direct damage it causes by sucking, ratherbecause it is the only known vector of the wheat dwarf virus(WDV) (Manurung et al., 2004; Praslicka, 1997). P. alienus is an oli-gophagous species. Its host plants are restricted to the Poaceaefamily (Lindblad and Areno, 2002; Manurung et al., 2005; Nickeland Remane, 2002), similarly to the host range of WDV (Mehneret al., 2003). Manurung et al. (2005) studied P. alienus populationdynamics in winter barley fields in Germany. They found that firstgeneration leafhoppers occurred in May/June developing onmaturing barley. Adults of the second generation first occurred inAugust, but the abundance of P. alienus did not peak until mid-Sep-tember on self-sown winter barley and stubble. They concludedthat self-sown barley fields are important in maintaining popula-tions of P. alienus. Other field studies on the spread of P. alienusand WDV in Sweden showed that catches of adults peaked atend of June during the occurrence of first generation, however,abundance of the second generation was lower, which could be aresult of lower monthly mean temperatures (Lindblad and Areno,2002). The Psammotettix genus has 20 species in Hungary, out ofwhich only P. alienus can be found in agricultural fields (Gy}orffyet al., 2009; Kiss et al., 2008). In Hungary P. alienus has two or threegenerations in a year; the leafhopper can be found from May untilthe end of November (Sáringer, 1989). In a previous study on P. ali-enus host range we tested host plant quality of 18 grass species andfound that most of the acceptable species are common in field mar-gins in Hungary (Tholt and Kiss, 2011). These studies show that fal-lows, pastures and other grassy areas act as important reservoirsfor the leafhopper and the virus (Lindblad and Areno, 2002; Meh-ner et al., 2003), and therefore any controlling factor that act inthese habitats are important both for the populations of P. alienusand for the virus it spreads.

Preliminary field studies showed that one potential predatoryspecies, that is likely to meet the criteria of co-occurring with P. ali-enus in space and time is the spider Tibellus oblongus (Walckenaer,1802) (Araneae, Philodromidae). Apart from generally knowing

that this is a fairly common species in agricultural habitat mosaics(Buchar and Ruzicka, 2002; Samu and Szinetár, 2002), we have hadvery little specific knowledge about the species. Spiders in the Phi-lodromidae family are known to hunt without web and to occurmainly above ground surface, in grass, on foliage and on bark;T. oblongus in particular ‘‘among grass’’ (Nentwig et al., 2013). Con-sidering the prey of T. oblongus there are two studies that reportthat the species might be important predator against Miridae bugs(Wegener, 1998) and thrips (Zrubecz et al., 2004).

Spiders are traditionally regarded as generalist predators, thatfeed on any prey that is suitable in size (Enders, 1975). However,field studies revealed that spiders, depending on taxon and forag-ing strategy, have very variable prey preferences (Nyffeler, 1999;Nyffeler et al., 1994). Toft and co-workers in a series of laboratoryexperiments proved that many common prey, such as aphids andcertain springtails, can be noxious to a number of agrobiont pred-atory arthropod species (Bilde and Toft, 1997; Toft, 1999). Proof ofconsumption and suitability may come from laboratory feedingexperiments (Mayntz and Toft, 2001).

In the present study aimed to establish whether T. oblongus ful-fills the main criteria of co-occurring with its potential prey inspace and time and feeding and developing on it. In particularwe intended to clarify the (i) spatial and (ii) temporal occurrenceof T. oblongus and the main factors that determine it; furtherwanted to know (iii) whether P. alienus is acceptable prey for thespider, and if yes, (iv) whether feeding on it is nutritionally ade-quate for growth and development.

2. Materials and methods

2.1. Field datasets

In studying the ecological characteristics of T. oblongus we se-lected datasets from the long term database of our group’s arach-nological research (Samu, 2000; Samu et al., 2008). Datasets,dated between 1994 and 2011, represented 1 year of collectingwith suction sampler in a habitat patch representative of the Hun-garian agricultural landscape mosaic. The habitat types were com-prised of (i) arable fields (cereals or alfalfa), (ii) grassy fieldmargins, (iii) disturbed, secondary grasslands and (iv) naturalgrassland patches (loess steppe remnants). We selected 96 data-sets from 62 habitat patches (some patches with 2 or 3 years ofdata), that represented 6 geographical regions scattered aboutthe non-montane agricultural areas of Hungary. To assess collect-ing method efficiency, further 24 datasets were considered fromthe same habitat patches where collecting was done using pitfalltraps. The datasets collectively contained data about 103,411 spi-der individuals (415 spider taxa), out of which 3528 belonged tothe genus Tibellus. The Tibellus genus has three species in Hungary,of which by far the most dominant is T. oblongus. Tibellus macellusSimon, 1875 is a faunistical rarity; Tibellus maritimus (Menge,1875) is also uncommon, preferring wetter habitats. At the 62 hab-itat patches considered not a single adult individual other thanT. oblongus was caught from the genus, thus all juvenile Tibellusspecimens could be safely regarded as T. oblongus.

2.2. Collecting methods

All suction sample data were collected by hand-held motorizedsuction sampler (Samu and Sárospataki, 1995), with a nozzle diam-eter of approx. 0.01 m2. 10 applications of the suction sampler in atransect consisted one sample (taken from an area of 0.1 m2). Theadditional method was pitfall trapping, which was carried out withplastic cups of 7 cm diameter at the opening, containing 40% ethyl-ene glycol as preservative mixed with a little amount of detergentto decrease surface tension.

392 F. Samu et al. / Biological Control 67 (2013) 390–396

P. alienus individuals for the laboratory experiments were col-lected by suction sampling, in the greatest numbers from a sownpasture near Kocs (47�35048.7400N, 18�14013.5800E) which was sup-plemented from the experimental fields of the Plant ProtectionInstitute near Nagykovácsi (47�32059.2900N, 18�55046.8400E). Wecollected T. oblongus individuals from field margins close to theseexperimental fields either by sweep net or suction sampler appli-cation. Previous long term monitoring indicated that T. oblonguswas the only Tibellus species occurring in the area.

2.3. Laboratory feeding experiments

After collection P. alienus specimens were transported to thelaboratory and placed onto potted young barley (Hordeum vulgarecv MV Jubilant); pots were covered with fine meshing. Pots werekept in a climate room with long daytime illumination (16/8 h,23/16 �C temperature regime). Tibellus individuals were kept inthe laboratory under the same conditions in individual plastic vialswith plaster of Paris bottom wetted for keeping humidity, and fedwith Drosophila prey twice a week.

Prey preference tests were performed in October 2011. Spidersreceived individual IDs. Hunger level was set uniform by providing5 Drosophila individuals on one day, and then starving the spidersfor 5 days. On the observation day spiders were weighed and in-spected for stage and condition. Freshly molted animals were ex-cluded. As the spiders were all juveniles, their sex could not bedetermined. Arenas were large Petri dishes (19 cm diam.) withplaster of Paris bottom and Plexiglas lid with a hole for the intro-duction of the animals (and plugged afterwards). Ten arenas wereobserved in parallel in a session for 70 min. Before starting theobservation 3 P. alienus and 3 Drosophila melanogaster individualswere introduced into each arena, then finally one spider. In total80 spiders were observed, timings and the subject of successfulpredations were noted.

The Tibellus growth experiments were conducted in two identi-cal bouts in March and May 2012, with 55 and 60 spiders, respec-tively. After identical basic feeding history (only Drosophila prey,no spiders reused from previous experiment) and weighing, spi-ders were assigned to one of four feeding regimes. In the respectivefeeding regimes (treatments) spiders received the followingamount and type of prey per feeding occasion: (a) 3 P. alienus;(b) 5 D. melanogaster; (c) 2 P. alienus + 2 D. melanogaster; and (d)starving control. The three feeding treatments, based on measure-ments of prey, were set to represent equal total fresh biomass ofprey. Feeding (after removing any remains) was done three timesduring the experiment, on days 0, 4 and 7. Spiders were weighedinitially to obtain a starting point, then on two more occasions,once in the middle of the feeding experiment, and once 4 days afterthe last feeding. Weighing was done on days 0, 7 and 11, beforefeeding if it was a feeding day (Fig. 5). The occurrence of moltswas noted. In both experimental bouts spiders were allocatedamong treatments after a randomization procedure which set themean and variation of the weight distributions nearly identicaland not significantly different from normal distribution. Animalswere not reused.

2.4. Data analysis

Field datasets were analyzed using percentage dominance (D%)of Tibellus as response variable. Here the maximum value was 40%,normal distribution was achieved with square root transformation.To evaluate responses of Tibellus D%, general linear mixed models(GLMM) were applied with year and region as random factors. Toassess prey choice we used the Manly–Chesson selectivity index(a) (Schmidt et al., 2012). For prey type 1 a1 = ln((n1 � r1)/n1)/[ln((n1 � r1)/n1) + ln((n2 � r2)/n2)], where n1 and n2 equal the initial

numbers, r1 and r2 are the numbers consumed of prey types 1and 2. Since aPsammotettix is not independent of aDrosophila, we onlytested for deviation in one index (aPsammotettix) from the hypotheti-cal equal choice value of a = 0.5. Time lag to attack the first preywas analyzed in a survival analysis framework, using parametricsurvival model with Weibull distribution, including censored cases(cases when no predation occurred, i.e. the ‘‘first prey’’ survived theobservation period). In the parametric model we considered preyidentity, spider weight and starvation length prior to experimentas explaining variables. Spider growth experiments were analyzedon spider weight as response variable using GLMM modelsor paired test, where the latter was appropriate. Initial modelselection showed no significant effect of experimental bouts, sothis effect was excluded from further models.

3. Results

3.1. The spatio-temporal distribution of T. oblongus

T. oblongus is a moderately dominant spider species in the agri-cultural landscape. Listing total catches and dominance values itwas the 5th in the order of spider taxa (Table 1). Pitfall traps area selective method for collecting spiders moving on the groundsurface, while suction sampling collects spiders vertically fromboth the ground surface and the vegetation above. Thus, if a speciesis found in suction samples and not in pitfall trap samples, thatindicates a preference for the stratum of the herbaceous vegeta-tion. D% ratio values indicated that preference for the herbaceouscanopy stratum was by an order of magnitude the highest in T.oblongus (Table 1).

Dominance of T. oblongus was significantly higher in the suctionsampling datasets as compared to pitfall trap datasets (GLMM,after controlling for year and region as random factors and habitattype as fixed factor, effect of the fixed factor sampling method:F = 42.68, d.f.=1, 107, P < 0.0001). Because of the extremely lowrepresentation of T. oblongus in the pitfall trap samples, all subse-quent analyses were performed on the 96 suction samplingdatasets.

T. oblongus had the highest dominance in the margins out of thefour main habitat types (GLMM, after controlling for year and re-gion as random factors, effect of habitat type as fixed factor:F = 8.93, d.f.=3, 77.2, P < 0.0001; with post hoc Tukey HSD test).Its mean dominance reached 13% there, which was significantlyhigher than dominance values in other habitats (Fig. 1).

Although suction sampling does not strictly allow to estimatepopulation density (Samu et al., 1997; Southwood and Henderson,2000), at least a rough idea about the density of the spiders can beobtained. Averaging suction sampling application over the 96 data-sets by habitats and converting figures to number of individuals/m2, the following figures were received (mean ± SD): disturbedgrassland 6.9 ± 12.26; arable field 2.2 ± 4.96; margin 10.9 ± 17.59;natural grassland 5.1 ± 9.49; all groups 5.1 ± 11.02. MaximumTibellus densities reached 100 individuals/m2 in certain margins.

Yearly changes are substantial in the T. oblongus population. Ifwe treat all variables as random factors in the statistical model thatdescribes Tibellus distribution, then we gain variance componentestimates. This estimation indicates that after habitat, which isresponsible for nearly 25% of variation in the data, yearly variationaccounts for c. 15% of the variance, while regional variation provedto be the smallest (Table 2).

Seasonal variation is also important if population interactionsare considered. To establish the life cycle of the species and thetiming of the population peak, we plotted monthly catches ofstages that we distinguished during identification (Fig. 2). T. oblon-gus overwinters in juvenile-subadult stage. First adults occur in

Table 1Total catch and dominance of the seven most dominant spider species in the agricultural landscape in the 96 suction sample datasets and their corresponding values in theadditional 24 pitfall trap datasets. Since in the case of Tibellus oblongus we considered both adults and juveniles, in other species congeneric juveniles were also consideredbelonging to the dominant species appearing in this table. That may slightly inflate these catches, but that makes for a conservative estimate regarding the dominance ofT. oblongus. Calculating only with adults gives the same rank order and very similar dominance values. D% = percentage dominance of T. oblongus in the total spider catch. D%ratio = D% suction/D% pitfall.

Spider taxon Suction sample data Pitfall trap data D% ratio

Catch D% Catch D%

Pardosa agrestis (Westring, 1861) 9140 11.0 11,208 55.1 0.2Meioneta rurestris (C. L. Koch, 1836) 5730 6.9 318 1.6 4.4Xysticus kochi Thorell, 1872 4853 5.8 423 2.1 2.8Pachygnatha degeeri Sundevall, 1830 3515 4.2 648 3.2 1.3Tibellus oblongus (Walckenaer, 1802) 3505 4.2 23 0.1 42Erigone dentipalpis (Wider, 1834) 2768 3.3 372 1.8 1.8Oedothorax apicatus (Blackwall, 1850) 1617 1.9 1254 6.2 0.3

Total spiders 83,077 20,334

Fig. 1. Dominance of Tibellus (% of total spider catch) in the main habitat types ofthe agricultural landscape. Natural grassland refers to loess steppe patches,disturbed grasslands are xerothermic and xeromesothermic secondary grassyareas. Dominance is statistically not different between habitats marked with thesame letter, as measured by Tukey HSD test at a = 0.05 level.

Table 2Restricted maximum likelihood estimation of variance components of the mixedmodel Sqrt (Tibellus D%) � intercept + rnd (Habitat) + rnd (Year) + rnd (Region).

Random effect Var component S.E. % of Total

Habitat 0.43 0.391 23.4Year 0.27 0.169 14.5Region 0.19 0.220 10.8Residual 0.94 0.149 51.3

Total 1.83 100

Fig. 2. Phenology of T. oblongus. Total monthly catches (summed across years) foreach stage category that were separated during identification are depicted.

Fig. 3. Prey preference of T. oblongus. Manly–Chesson selectivity index for the twoprey types based on 80 choice trials. The line at y = 0.5 signifies the equal choicevalue.

F. Samu et al. / Biological Control 67 (2013) 390–396 393

April and can be caught until autumn, although their numbers startto decrease after June. Reproduction starts in May and spiderlings(1st stage instars) emerge from June. Numerically the largest num-bers of T. oblongus were caught between May and October, mostlyrepresented by juvenile stages.

3.2. Prey preference experiments

In the prey choice experiment 82.5% of the 80 spiders preyed onat least one prey item. Six spiders (7.5%) preyed twice within theobservation period. Choice between Psammotettix and Drosophilawas very close to equal, out of the 66 preying spiders 34 chose leaf-hopper. That translated to a Manly–Chesson selectivity index valueof aPsammotettix = 0.54 ± 0.059 (mean ± S.E.) (Fig. 3), a value that was

not significantly different from the hypothetical value of 0.5, signi-fying equal choice (t = 0.64, d.f.=65, P = 0.53).

It was also considered how soon attacking the prey occurred inthe choice arenas. Since preying on a second prey item had a verylow occurrence, we considered time lag only until first predation,and treated cases where no predation occurred as censored. The re-sults of the parametric survival model showed that neither spiderweight (v2 = 0.175; P = 0.67), nor prey identity (v2 = 2.439;P = 0.12) was a significant factor; i.e. lag to attack did not differ

Fig. 5. Weight changes of spiders in the feeding experiments by treatment groups.Changes are shown by fitted linear regression lines. Arrows indicate the timing offeeding.

394 F. Samu et al. / Biological Control 67 (2013) 390–396

between prey types (Fig. 4) and was not dependent on spider size.The only factor significantly affecting how soon predation occurredwas the hunger state of spiders (v2 = 6.547; P = 0.01).

3.3. The suitability of Psammotettix as prey for Tibellus

During the short term feeding experiment we tested whetherthere was a difference between weight gain of spiders in the differ-ent feeding regimes. Final weight showed a significant increasecompared to the initial weight in all three fed groups (GLMM onspider weight, with spiders as random factors, feeding period andfeeding regime fixed factors; effect of feeding period (initial vs. fi-nal): F = 147.41; d.f.=1, 91; P < 0.0001; effect of prey type:F = 0.149; d.f.=2, 89; P = 0.861), while restricting statistical testingto the single group of starving spiders (control), weight showed asignificant decline (paired t-test between initial and final weight:t = �3.27; d.f.=30; P < 0.003). Considering only the fed groups andthe three weight measurement days as a continuous variable, usingthe above modeling approach, we get similar results: there is ahighly significant change over time (F = 242.4; d.f.=1, 183;P < 0.0001), but there is no significant effect of the feeding regime(F = 0.078; d.f.=2, 89; P = 0.924) (Fig. 5).

4. Discussion

Agrobiont species are the ‘‘super dominant species’’ of the agri-cultural assemblages (Luczak, 1979; Samu and Szinetár, 2002). Alist of the Hungarian agrobiont spider species has been maintainedbased on our long term projects (Samu and Szinetár, 2002). T.oblongus was the 6th on the 2002 list (only adult data considered),which was based both on suction and pitfall trap samples. Thus,the presence of T. oblongus has been known, and was not verymuch different from what we report here based on data that in-cludes juveniles and recent years’ data, as well. So, why do we at-tach more significance to this species now, and think that it can bean important actor in pest control? The answer lies in a broaderinvestigation of the species habitats, in a trait-based assessmentof Tibellus and other agrobiont species and the eventual feedingexperiments we conducted.

We think the present Tibellus case study shows, that natural en-emy candidates are not necessarily species that have the highestdominance in the fields, but could also be species that maintainhigh abundance in neighboring habitats. Landscape elements,other than the agricultural fields themselves, have been increas-ingly recognized to play role in the maintenance of biodiversity,

Fig. 4. The cumulative proportion of the first prey remaining to be attacked by preytype, considering all cases where successful predation occurred (i.e. the uncensoredcases are depicted).

pest control and other ecosystem services, such as pollination(Lovei et al., 2006; Oberg et al., 2007; Westphal et al., 2006). Bian-chi et al. (2006) asserts that landscape-driven pest suppressionoccurred in complex landscapes with herbaceous habitats (e.g.fallows, field margins) in 80% of the reviewed cases. However,other studies indicate that the positive effect of margins dimin-ishes with distance (Samu et al., 1999). Thus, both the ratio of nat-ural elements in a landscape and their configuration are importantfactors of biodiversity and natural pest control (Fahrig et al., 2011).Margins are not only a common semi-natural element of the agri-cultural landscape, but are also ecotone habitats where abundanceand diversity may peak. Edge effects may take various forms (Dan-gerfield et al., 2003; Ries and Sisk, 2004; Walker et al., 2003), but itis not uncommon that – like we found with Tibellus – arthropodpredators are concentrated at edges (Girma et al., 2000; Loveiet al., 2006; Muff et al., 2009). Arable – grassy habitat edges mayrepresent such ecotones where an increase in richness and abun-dance can be expected, because the two neighboring habitats con-tain different resources (e.g. different prey animals) (Ries and Sisk,2004). To translate such a scenario to the present field situation,we might speculate that Tibellus has an increased abundance atthe edges, because it is their preferential grassy habitat, and canstill enjoy prey from both the arable field and the grassland.

If pest species need a secondary habitat either because of theirlife cycle or because of other demands, the importance of naturalenemies in those habitats is especially high. Such a cycle of coloni-zation and recolonization is known for a wide range of pests (Bah-lai et al., 2010; Frouz and Paoletti, 2000; Huseth et al., 2012), aprocess that might be followed rapidly enough by the migrationof natural enemies (Sivakoff et al., 2012). Some modeling studiesgo that far that they suggest to increase field size in order to de-crease pest immigration from neighboring habitats (Segoli andRosenheim, 2012). Leafhoppers are known to switch hosts and dis-perse between habitat patches (Lamp et al., 1994; Novotny, 1994).P. alienus, our focal leafhopper pest, is also known for host switch-ing and associated dispersal (Lindblad and Sigvald, 2004). There-fore, natural enemies residing in its non-crop habitats, such asfallows, leys and permanent pastures (Lindblad and Areno, 2002)can be important factors in controlling the immigration of the spe-cies into autumn cereals (Lindblad and Waern, 2002; Praslicka,1997).

Apart from being in either the primary or secondary habitat ofpests, the effectiveness of spiders as natural enemies is also deter-mined whether they live in the right stratum or microhabitat.A majority of the agrobiont spider species that we found in

F. Samu et al. / Biological Control 67 (2013) 390–396 395

Hungary (Samu and Szinetár, 2002) live either on or close to theground surface. In the case of Linyphiidae, which is overall the mostabundant spider family in agricultural surveys, webs are built inthe lower stratum of the fields, among the stems of cereals or evenattached to rough soil (Harwood et al., 2003; Sunderland et al.,1986). Spiders on the ground surface may have important role incapturing herbivorous pest insects that move from one plant to an-other or are dislodged by disturbance (Sunderland, 1999). Still, it isnatural that spiders which are the closest to the pests can have thehighest impact on them (Nentwig and Heimer, 1987; Nyffeler,1999; Rypstra, 1982). Tibellus in this respect was the most abun-dant species that foraged in the grass canopy stratum, where Psam-motettix feeds. Such proximity may have further effects on theleafhoppers. Foraging by spiders at the feeding place of herbivorescaused indirect effects on plants by changing grasshopper foragingbehavior (Schmitz et al., 1997). Meta-analysis showed that micro-habitat and predator hunting modes may alter prey behaviorthrough non-consumptive effects, which can have ecosystem levelconsequences (Hawlena and Schmitz, 2010; Preisser et al., 2007).

To assess whether a predator species may be a successful natu-ral enemy of a certain pest, last but not least, we should establishwhether the prey is among the preferential prey types, and if feed-ing on it has no toxic effect, further, prey nutrients allow growthand development of the predator. Common pest species, especiallyaphids turned out to have toxic or noxious effect on predatorswhich were previously regarded as their natural enemy (Bildeand Toft, 2001). For instance Pardosa amentata and Erigone atrajuveniles were unable to develop on a pure diet of the aphid Rho-palosiphum padi; they died without molting (Toft, 1995). The pres-ent experiments were important in this regard to prove thatP.alienus is clearly among the preferred prey types for Tibellus,and even prolonged feeding has no adverse effect on the develop-ment of the spiders.

5. Conclusions

The present paper collected circumstantial ecological evidencefor the co-occurrence of a pest and its suspected biological controlagent. T. oblongus has substantial presence in key habitats andmicrohabitats of the leafhopper pest and the seasonal overlap isalso nearly complete. To obtain more than anecdotal prey spec-trum information for spiders directly in the field is extremely dif-ficult (as an example, see Nyffeler et al., 1994), therefore weconducted laboratory prey choice experiments, to establish thatP. alienus is among the preferred prey types of Tibellus. However,these investigations must be only a starting point. We will needDNA gut analysis (Kerzicnik et al., 2012) to provide direct evidencefor consumption, and further field studies to quantify the signifi-cance of predation.

Acknowledgments

We thank Erika Botos for help in collection, selection and iden-tification of samples. This project was funded by OTKA K81971.

References

Bahlai, C.A., Sikkema, S., Hallett, R.H., Newman, J., Schaafsma, A.W., 2010. Modelingdistribution and abundance of soybean aphid in soybean fields usingmeasurements from the surrounding landscape. Environmental Entomology39, 50–56.

Bianchi, F., Booij, C.J.H., Tscharntke, T., 2006. Sustainable pest regulation inagricultural landscapes: a review on landscape composition, biodiversity andnatural pest control. Proceedings of the Royal Society B – Biological Sciences273, 1715–1727.

Bilde, T., Toft, S., 1997. Limited predation capacity by generalist arthropod predatorson the cereal aphid, Rhopalosiphum padi. Biological Agriculture & Horticulture15, 143–150.

Bilde, T., Toft, S., 2001. The value of three cereal aphid species as food for ageneralist predator. Physiological Entomology 26, 58–68.

Buchar, J., Ruzicka, V., 2002. Catalogue of Spiders of the Czech Republic. Peres,Praha.

Christian, M.L., Willis, W.G., 1993. Survival of wheat streak moasic-virus in grasshosts in Kansas from wheat harvest to fall wheat emergence. Plant Disease 77,239–242.

Dangerfield, J.M., Pik, A.J., Britton, D., Holmes, A., Gillings, M., Oliver, I., Briscoe, D.,Beattie, A.J., 2003. Patterns of invertebrate biodiversity across a natural edge.Austral Ecology 28, 227–236.

Enders, F., 1975. The influence of hunting manner on prey size, particularly inspiders with short attack distances. American Naturalist 109, 737–763.

Fahrig, L., Baudry, J., Brotons, L., Burel, F.G., Crist, T.O., Fuller, R.J., Sirami, C.,Siriwardena, G.M., Martin, J.L., 2011. Functional landscape heterogeneityand animal biodiversity in agricultural landscapes. Ecology Letters 14, 101–112.

Frouz, J., Paoletti, M.G., 2000. Spatial distribution of different life stages of oneDipteran community along hedgerow and field margin. Landscape and UrbanPlanning 48, 19–29.

Girma, H., Rao, M.R., Sithanantham, S., 2000. Insect pests and beneficial arthropodspopulation under different hedgerow intercropping systems in semiarid Kenya.Agroforestry Systems 50, 279–292.

Gy}orffy, G., Kiss, B., Koczor, S., Orosz, A., 2009. Hemiptera: Archaeorrhyncha,Clypeorrhyncha. Hungarian Natural History Museum, Budapest.

Harwood, J.D., Sunderland, K.D., Symondson, W.O.C., 2003. Web-location bylinyphiid spiders: prey-specific aggregation and foraging strategies. Journal ofAnimal Ecology 72, 745–756.

Hawlena, D., Schmitz, O.J., 2010. Physiological stress as a fundamentalmechanism linking predation to ecosystem functioning. American Naturalist176, 537–556.

Huberty, A.F., Denno, R.F., 2004. Plant water stress and its consequences forherbivorous insects: a new synthesis. Ecology 85, 1383–1398.

Huseth, A.S., Frost, K.E., Knuteson, D.L., Wyman, J.A., Groves, R.L., 2012. Effects oflandscape composition and rotation distance on Leptinotarsa decemlineata(Coleoptera: Chrysomelidae) abundance in cultivated potato. EnvironmentalEntomology 41, 1553–1564.

Kerzicnik, L.M., Chapman, E.G., Harwood, J.D., Peairs, F.B., Cushing, P.E., 2012.Molecular characterization of Russian wheat aphid consumption by spiders inwinter wheat. Journal of Arachnology 40, 71–77.

Kindler, S.D., Jensen, K.B., Springer, T.L., 1999. Detection and characterization ofmechanisms of resistance to Russian wheat aphid (Homoptera: Aphididae) incrested wheatgrass. Journal of Agricultural and Urban Entomology 16, 129–139.

Kiss, B., Rédei, D., Koczor, S., 2008. Occurrence and feeding of hemipterans oncommon ragweed (Ambrosia artemisiifolia) in Hungary. Bulletin of Insectology61, 195–196.

Lamp, W.O., Nielsen, G.R., Danielson, S.D., 1994. Patterns among host plants ofpotato leafhopper, Empoasca fabae (Homoptera, Cicadellidae). Journal of theKansas Entomological Society 67, 354–368.

Lindblad, M., Areno, P., 2002. Temporal and spatial population dynamics ofPsammotettix alienus, a vector of wheat dwarf virus. International Journal ofPest Management 48, 233–238.

Lindblad, M., Sigvald, R., 2004. Temporal spread of wheat dwarf virus and matureplant resistance in winter wheat. Crop Protection 23, 229–234.

Lindblad, M., Waern, P., 2002. Correlation of wheat dwarf incidence to winter wheatcultivation practices. Agriculture Ecosystems & Environment 92, 115–122.

Lovei, G.L., Magura, T., Tothmeresz, B., Kodobocz, V., 2006. The influence of matrixand edges on species richness patterns of ground beetles (Coleoptera:Carabidae) in habitat islands. Global Ecology and Biogeography 15, 283–289.

Luczak, J., 1979. Spiders in agrocoenoses. Polish Ecological Studies 5, 151–200.Manurung, B., Witsack, W., Mehner, S., Gruntzig, M., Fuchs, E., 2004. The

epidemiology of wheat dwarf virus in relation to occurrence of the leafhopperPsammotettix alienus in Middle-Germany. Virus Research 100, 109–113.

Manurung, B., Witsack, W., Mehner, S., Gruntzig, M., Fuchs, E., 2005. Studies onbiology and population dynamics of the leafhopper Psammotettix alienus Dahlb.(Homoptera: Auchenorrhyncha) as vector of wheat dwarf virus (WDV) inSaxony-Anhalt, Germany. Zeitschrift Fur Pflanzenkrankheiten UndPflanzenschutz-Journal of Plant Diseases and Protection 112, 497–507.

Mayntz, D., Toft, S., 2001. Nutrient composition of the prey’s diet affects growth andsurvivorship of a generalist predator. Oecologia 127, 207–213.

Mehner, S., Manurung, B., Gruntzig, M., Habekuss, A., Witsack, W., Fuchs, E., 2003.Investigations into the ecology of the wheat dwarf virus (WDV) in Saxony-Anhalt, Germany. Zeitschrift Fur Pflanzenkrankheiten Und Pflanzenschutz-Journal of Plant Diseases and Protection 110, 313–323.

Muff, P., Kropf, C., Frick, H., Nentwig, W., Schmidt-Entling, M.H., 2009. Co-existenceof divergent communities at natural boundaries: spider (Arachnida: Araneae)diversity across an alpine timberline. Insect Conservation and Diversity 2, 36–44.

Nentwig, W., Heimer, S., 1987. Ecological aspects of spider webs. In: Nentwig, W.(Ed.), Ecophysiology of Spiders. Springer-Verlag, Berlin, pp. 211–225.

Nentwig, W., Blick, T., Gloor, D., Hänggi, A., Kropf, C., 2013. Spiders of Europe.Available from: <http://www.araneae.unibe.ch> (version 2.2013).

Nickel, H., Remane, R., 2002. Artenliste der Zikaden Deutschlands, mit Angabe vonNährpflanzen, Nahrungsbreite, Lebenszyklus, Areal und Gefährdung (Hemiptera,Fulgoromorpha et Cicadomorpha) [Check list of the planthoppers andleafhoppers of Germany, with notes on food plants, diet width, life cycles,geographic range and conservation status]. Beiträge zur Zikadenkunde 5, 27–64.

396 F. Samu et al. / Biological Control 67 (2013) 390–396

Novotny, V., 1994. Association of polyphagy in leafhoppers (Auchenorrhyncha,Hemiptera) with unpredictable environments. Oikos 70, 223–232.

Nyffeler, M., 1999. Prey selection of spiders in the field. Journal of Arachnology 27,317–324.

Nyffeler, M., Sunderland, K.D., 2003. Composition, abundance and pest controlpotential of spider communities in agroecosystems: a comparison of Europeanand US studies. Agriculture Ecosystems & Environment 95, 579–612.

Nyffeler, M., Sterling, W., Dean, D., 1994. How spiders make a living. EnvironmentalEntomology 23, 1357–1367.

Oberg, S., Ekbom, B., Bommarco, R., 2007. Influence of habitat type and surroundinglandscape on spider diversity in Swedish agroecosystems. AgricultureEcosystems & Environment 122, 211–219.

Praslicka, J., 1997. Influence of occurrence of Psammotettix alienus on the degree ofattack of winter wheat with wheat dwarf virus (WDV). Rostlinna Vyroba 43,81–85.

Preisser, E.L., Orrock, J.L., Schmitz, O.J., 2007. Predator hunting mode and habitatdomain alter nonconsumptive effects in predator–prey interactions. Ecology 88,2744–2751.

Riechert, S.E., Lockley, T., 1984. Spiders as biological control agents. Annual Reviewof Entomology 29, 299–320.

Ries, L., Sisk, T.D., 2004. A predictive model of edge effects. Ecology 85, 2917–2926.Rypstra, A.L., 1982. Building a better insect trap – an experimental investigation of

prey capture in a variety of spider webs. Oecologia 52, 31–36.Rypstra, A.L., Marshall, S.D., 2005. Augmentation of soil detritus affects the spider

community and herbivory in a soybean agroecosystem. EntomologiaExperimentalis et Applicata 116, 149–157.

Samu, F., 2000. A general data model for databases in experimental animal ecology.Acta Zoologica Academiae Scientiarum Hungaricae 45, 273–292.

Samu, F., Sárospataki, M., 1995. Design and use of a hand-hold suction sampler andits comparison with sweep net and pitfall trap sampling. Folia EntomologicaHungarica 56, 195–203.

Samu, F., Szinetár, C., 2002. On the nature of agrobiont spiders. Journal ofArachnology 30, 389–402.

Samu, F., Németh, J., Kiss, B., 1997. Assessment of the efficiency of a hand-heldsuction device for sampling spiders: improved density estimation oroversampling? Annals of Applied Biology 130, 371–378.

Samu, F., Sunderland, K.D., Szinetar, C., 1999. Scale-dependent dispersal anddistribution patterns of spiders in agricultural systems: a review. Journal ofArachnology 27, 325–332.

Samu, F., Csontos, P., Szinetar, C., 2008. From multi-criteria approach to simpleprotocol: assessing habitat patches for conservation value using species rarity.Biological Conservation 141, 1310–1320.

Sandstrom, J., 2000. Nutritional quality of phloem sap in relation to host plant-alternation in the bird cherry-oat aphid. Chemoecology 10, 17–24.

Sáringer, G., 1989. Egyenl}o szárnyú rovarok – Homoptera. Kabócák –Auchenorrhyncha [Homopteran insects – Auchenorrhyncha]. In: Jermy, T.,Balázs, K. (Eds.), A növényvédelmi állattan kézikönyve [Handbook of Zoology inPlant Protection]. Akadémiai Kiadó, Budapest, pp. 13–75.

Schmidt, J.M., Sebastian, P., Wilder, S.M., Rypstra, A.L., 2012. The nutritional contentof prey affects the foraging of a generalist arthropod predator. PLoS One 7,e49223.

Schmitz, O.J., Beckerman, A.P., Obrien, K.M., 1997. Behaviorally mediated trophiccascades: effects of predation risk on food web interactions. Ecology 78, 1388–1399.

Segoli, M., Rosenheim, J.A., 2012. Should increasing the field size of monoculturalcrops be expected to exacerbate pest damage? Agriculture Ecosystems &Environment 150, 38–44.

Sivakoff, F.S., Rosenheim, J.A., Hagler, J.R., 2012. Relative dispersal ability of a keyagricultural pest and its predators in an annual agroecosystem. BiologicalControl 63, 296–303.

Southwood, T.R.E., Henderson, P.A., 2000. Ecological Methods. Blackwell Science,Oxford.

Sunderland, K., 1999. Mechanisms underlying the effects of spiders on pestpopulations. Journal of Arachnology 27, 308–316.

Sunderland, K.D., Fraser, A.M., Dixon, A.F.G., 1986. Distribution of linyphiid spidersin relation to capture of prey in cereal fields. Pedobiologia 29, 367–375.

Tholt, G., Kiss, B., 2011. A csíkos gabonakabóca tápnövénykörének vizsgálata. [Hostrange of Psammotettix alienus]. Növényvédelem 47, 229–235.

Toft, S., 1995. Value of the aphid Rhopalosiphum padi as food for cereal spiders.Journal of Applied Ecology 32, 552–560.

Toft, S., 1999. Prey choice and spider fitness. Journal of Arachnology 27, 301–307.Walker, S., Wilson, J.B., Steel, J.B., Rapson, G.L., Smith, B., King, W.M., Cottam, Y.H.,

2003. Properties of ecotones: evidence from five ecotones objectivelydetermined from a coastal vegetation gradient. Journal of Vegetation Science14, 579–590.

Wegener, C., 1998. Predation on the grassbug species Notostira elongata(Heteroptera: Miridae) by Nabidae (Heteroptera) and selected non-webbuilding spiders (Araneae). Entomologia Generalis 22, 295–304.

Welch, K.D., Crain, P.R., Harwood, J.D., 2011. Phenological dynamics of web-buildingspider populations in alfalfa: implications for biological control. Journal ofArachnology 39, 244–249.

Westphal, C., Steffan-Dewenter, I., Tscharntke, T., 2006. Bumblebees experiencelandscapes at different spatial scales: possible implications for coexistence.Oecologia 149, 289–300.

Williams, C.T., 1995. Effects of plant age, leaf age and virus yellows infection on thepopulation dynamics of Myzus persicae (Homoptera: Aphididae) on sugarbeet infield plots. Bulletin of Entomological Research 85, 557–567.

Wissinger, S., 1997. Cyclic colonization in predictably ephemeral habitats: atemplate for biological control in annual crop systems. Biological Control 10,4–15.

Zrubecz, P., Tóth, F., Nagy, A., 2004. Pókfajok (Xysticus kochi Thorell; Tibellusoblongus Walckenaer) lárváinak hatékonyságvizsgálata virágtripszek(Frankliniela spp.) elen hajtatott paprikában [Study into the effectiveness oflarvae of spider species (Xysticus kochi Thorell; Tibellus oblongus Walckenaer)against flower thrips (Frankliniela spp.) in greenhouse pepper]. Növényvédelem40, 527–533.