ZOOLOGY Zoology 111 (2008) 2–8

Author's personal copy

ARTICLE IN PRESS ZOOLOGY Zoology 111 (2008) 2–8 www.elsevier.de/zool

Orientation of the pit-building antlion larva Euroleon (Neuroptera, Myrmeleontidae) to the direction of substrate vibrations caused by prey$ Bojana Mencinger-Vracˇko, Dusˇan DevetakÃ

Department of Biology, Faculty of Natural Sciences and Mathematics, University of Maribor, Korosˇka 160, 2000 Maribor, Slovenia

Received 13 March 2007; received in revised form 8 May 2007; accepted 10 May 2007

Abstract

Pit-building antlion Euroleon nostras constructs efﬁcient traps in sand to catch its prey. The predator is known to react to substrate vibrations produced by movements of its prey outside the pit with sand-tossing behaviour but it has not yet been ascertained if this reaction is directed towards the prey. The accuracy of the sand-tossing response in the presence of four prey species was measured using a video recording method. The sand-tossing angle was highly positively correlated with the prey angle. Sand tossing was most frequently elicited when prey was on the posterior sand surface. Covering the larval photoreceptors did not inﬂuence the antlion’s localizing behaviour. r 2007 Elsevier GmbH. All rights reserved.

Keywords: Antlion; Myrmeleontidae; Predatory behaviour; Sand tossing

Introduction avalanches typical of crater angles (Fertin and Casas 2006). The larvae of most antlion species (Neuroptera: When the prey arrives at the bottom of the trap, the Myrmeleontidae) are sand-dwelling insects, but only a larva rapidly grasps it with its mandibles. If the antlion few of them construct conical pits in dry, loose sand to does not succeed in catching the prey at its first attempt capture prey (for reviews see Gepp and Ho¨lzel 1989; or if the prey evades the antlion and tries to climb up the Scharf and Ovadia 2006; Devetak et al. 2007a). While walls of the trap, the larva tosses sand with violent flicks antlion larvae feed on a variety of arthropods, ants of its head and mandibles until the prey slides back to usually constitute the majority of prey items (Topoff the antlion’s jaws (Topoff 1977; Griffiths 1980). 1977; Griffiths 1980; Gepp and Ho¨lzel 1989). The pit The function of pit-building is obvious. The pit functions by conveying the prey towards the pit centre funnels prey to the antlion’s mandibles, and it also (Lucas 1982). The pit is an efficient trap, with slopes retards the escape of prey and consequently increases steep enough to guide prey to the mandibles without any the amount of time the prey is available for capture. attack, and shallow enough to avoid the likelihood of According to Mansell (1996, 1999), the advantages of constructing a pitfall trap as a predation strategy also include: (i) the need to hunt or pursue prey is reduced, $This paper is dedicated to Matija Gogala on the occasion of his 70th birthday. thereby conserving energy; (ii) the pit is a selective device ÃCorresponding author. Tel.: +386 22282160. for prey of a suitable size, because large prey would be E-mail address: [email protected] (D. Devetak). able to escape, and energy is not wasted on unsuccessful

ARTICLE IN PRESS B. Mencinger-Vracˇko, D. Devetak / Zoology 111 (2008) 2–8 3 attacks; (iii) it affords protection as large species falling adults of woodlice Trachelipus rathkei Brandt, ant into the pit cause the larva to retreat; (iv) fast-moving Formica sp., firebug Pyrrhocoris apterus (L.) and meal- prey can be intercepted by the pit; (v) prey can be worm beetle Tenebrio molitor L. rapidly subdued as it is disorientated upon falling into To reduce vibratory noise from the surroundings the the pit. A pit retards prey escape regardless of the plastic container with sand was placed on a sand layer presence or absence of an antlion (Devetak 2005). and this rested on cork, a mineral-wool layer and a The antlion detects its prey from a distance of a few concrete plate (75 50 4cm3, 37 kg) supported by a centimetres by sensing the vibrations that the prey mineral-wool layer. In addition, the whole setup was produces during locomotion (Devetak 1985; Mencinger placed on a vibration-free table. All measurements were 1998; Devetak et al. 2007b). It is known that sand- done in an anechoic chamber (130 90 160 cm3). tossing behaviour occurs during construction of the pit A prey item was carefully placed on the sand surface and when the prey is already trapped in the pit and tries 20 cm away from the centre of the pit and the activity of to evade the predator (Youthed and Moran 1969; the prey and the predator-prey interaction was video- Topoff 1977; Griffiths 1980). In this study sand-tossing taped at a distance of ca. 70 cm. Recording commenced behaviour of the European antlion species Euroleon with prey introduction and ended when the prey either nostras (Geoffroy in Fourcroy, 1785) is determined as a escaped from the area, or was consumed. Trials were response to prey prior to entering the pit. The larva evaluated only if the antlion responded to the presence reacts to the presence of prey approaching the pit rim of a prey item within 2 min. without seeing it. The aim of our study is to answer the The direction in which the head of the antlion was question whether this sand-tossing reaction is directed pointing is indicated as 01, and the direction of the end toward the prey or not. of the abdomen as 1801 (Fig. 1). At the moment when the larva reacted to the presence of a prey item with sand tossing, we measured two azimuth angles, the angle between the antlion’s long axis and the position of the Materials and methods prey (prey angle), and the angle between the antlion’s long axis and the end point of the sand toss (sand- Third-instar larvae of E. nostras collected in the tossing angle). The sand area in front of the head with surroundings of Maribor, Slovenia, were used in the jaws was defined as the anterior sand surface and the present study. Larval stages were determined by mea- sand area behind the caudal/dorsal part of the antlion’s suring head capsule width and body length (Devetak body as the posterior sand surface. Additionally, we et al. 2005). measured the distance between the midpoint of the prey Antlion larvae were kept in the laboratory at room and the centre of the pit. As larvae reacted at different temperature (22–26 1C) in sieved sand. In all experi- distances, only the maximum distance was evaluated ments fine-grained quartzite sand (Kema Puconci d.d.) during each trial. The depth and diameter of the pit were with the following grain size (gs) distribution was used: gs o0.1 mm, 15.5 weight% (wt%); gs 0.1–0.25 mm, 32.9 wt%; gs 0.25–0.315 mm, 15.2 wt%; gs 0.315– 0.5 mm, 29 wt%; gs 0.5–1 mm, 4.4 wt%; gs 1–1.5 mm, 3 wt% (mechanical analysis with sieves according to DIN 1170 and DIN 1171). In a previous study (Devetak et al. 2007b) it was shown that this kind of sand is most convenient for successful prey-catching behaviour in antlions. The moisture content of the sand was less than 2% by weight. Before experimental treatment the larvae were kept singly in plastic containers (25 20 5.5 cm3) filled with sand. Workers of the ant species Lasius fuliginosus (Latreille, 1798) and L. emarginatus (Olivier, 1791) were used as food source for the antlions. Feeding took place every day and one ant was delivered to each pit. Antlions used in the experiments were placed singly into a plastic container (40 30 5.5 cm3) filled with sand to a depth of 5 cm.Measurement started a day after the larva had constructed a pit in the centre of the Fig. 1. Sand-tossing behaviour of Euroleon nostras in the container. All prey animals were adults and laboratory- presence of prey. The antlion is positioned in the middle of the bred. The prey animals used in the experiments were bottom of the pit, and the prey is located outside the pit. Author's personal copy

ARTICLE IN PRESS 4 B. Mencinger-Vracˇko, D. Devetak / Zoology 111 (2008) 2–8 also determined. From the data obtained we calculated from a PC monitor. Trials were not evaluated when a reaction distance, i.e. the distance between the prey and prey item was moving at the rim of the pit. The accuracy the antlion sitting at the bottom of the pit (see Devetak of the sand-tossing response was measured from selected 1985). frames using the computer program Mirror Video- Details of the antlion’s sand-tossing behaviour were Analysis for Windows. Selected frames were super- recorded on video tape and then traced frame-by-frame imposed on each other and prey and sand-tossing

0° 0° 0° 0°

270° 90° 270° 90° 270° 90° 270° 90°

180° 180° 180° 180° Tenebrio molitor Trachelipus rathkei

350 350 300 300 250 250 200 200 150 150 100 100 50 50 Sand tossing angle (degrees)

0 Sand tossing angle (degrees) 0

0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350 Prey angle (degrees) Prey angle (degrees)

0° 0° 0° 0°

270° 90° 270° 90° 270° 90° 270° 90°

180° 180° 180° 180° Formica sp. Pyrrhocoris apterus 350 350 300 300 250 250 200 200 150 150 100 100 50 50 Sand tossing angle (degrees) Sand tossing angle (degrees) 0 0 0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350 Prey angle (degrees) Prey angle (degrees)

Figs. 2–5. 2. Accuracy of sand-tossing response of unimpeded antlions in the presence of prey. Fig. 2: mealworm Tenebrio molitor (N ¼ 91, n ¼ 337). 3. woodlouse Trachelipus rathkei (N ¼ 23, n ¼ 87). 4. ﬁrebug Pyrrhocoris apterus (N ¼ 10, n ¼ 28). 5. ant Formica sp. (N ¼ 11, n ¼ 21). N – number of antlions, n – number of trials. (a) Prey angle at the moment when sand tossing occurred. (b) Sand-tossing angle. Each sign in (a) and (b) represents a single response (except in Fig. 2 at the angle of 1801 where it represents 4 responses). (c) Relationship between prey angle and sand-tossing angle. Author's personal copy

ARTICLE IN PRESS B. Mencinger-Vracˇko, D. Devetak / Zoology 111 (2008) 2–8 5 angles were determined for a single response at 51 precision. The relationship between target and sand-tossing angles was determined with correlation analysis and Watson’s F-test for circular mean directions with the computer programs Oriana for Windows 1.0 (Batschelet 1981; Fisher 1995) and Statistica 5.1 (Sokal and Rohlf 1995). Two sets of experiments were conducted with mealworms – an experiment with unimpeded antlions and an experiment with antlion larvae with covered eyes. To exclude vision all stemmata positioned on the eye tubercles were covered with a non-transparent paint.

Results

Reaction distance

Movement of a prey item evoked sand tossing and reaction distance was calculated as the distance between Fig. 6. Accuracy of sand-tossing response of antlions with the antlion and its prey at the moment when the reaction covered eyes in the presence of a mealworm, Tenebrio molitor was elicited. The mean reaction distances (7SD) were (N ¼ 32, n ¼ 101). N – number of antlions, n – number of 8.471.4 cm (ant Formica; number of measurements, trials. (a) Prey angle at the moment when sand tossing n ¼ 19), 8.771.9 cm (firebug Pyrrhocoris; n ¼ 21), occurred. (b) Sand-tossing angle. Each sign in (a) and (b) 9.172.3 cm (woodlouse Trachelipus; n ¼ 35), and 10.17 represents a single response. (c) Relationship between prey 2.7 cm (mealworm Tenebrio; n ¼ 30). As the F-test angle and sand-tossing angle. revealed, reaction distances differed significantly for different prey (F4, 149 ¼ 2.46; P ¼ 0.0470). The sand-tossing angle was highly positively correlated with the prey angle and there was a significant relationship between the variables (Figs. 2c and 6c). Accuracy of prey localization The correlation coefficient varied from r ¼ 0.78 (for Trachelipus, Po0.001) to r ¼ 0.88 (for Formica, Antlions responded to prey near the pit with opening Po0.001; Table 1). the jaws, sand tossing and translocation in the pit. Sand Since among all four prey species antlions most tossing occurred frequently when prey was on the frequently responded to mealworms (Tenebrio), this posterior and posterolateral sand surface outside the prey species was chosen to assay the antlion’s respon- pit (Figs. 2a and 6a). The antlion larvae tossed sand in siveness under modified experimental conditions. When the direction of a prey item without seeing it. When a the antlion’s eyes were covered the sand-tossing angle prey item was not on the antlion’s posterior side the was still highly positively correlated with the prey angle antlion larva often waited with its jaws agape. (Fig. 6), with the correlation coefficient (r ¼ 0.80; When prey approached the antlion from the anterior Po0.001) similar to that of unimpeded antlions or anterolateral direction, the larva often changed its (r ¼ 0.79; Po0.001). position so that the prey became positioned posteriorly, and started sand tossing. During the experiments the antlion larvae changed their orientation within the pit several times. Only in a few cases the antlions moved up Discussion from the bottom of the pit to the pit rim. To assess how accurately antlions determine prey Antlion larvae are psammophilous insects which direction, the prey angle was compared with the sand- identify and localize their prey by means of substrate tossing angle at the moment when sand tossing vibrations (Devetak 1985; Devetak et al. 2007b). Dry occurred. As Watson’s F-test for circular mean direc- sand provides sufficient information to detect prey tions revealed, the differences between the prey angles of (Brownell 1977; Brownell and Farley 1979a, b). four prey species and the sand-tossing angles were Antlion larvae require dry substrates with a certain statistically not significant (Table 1). particle size (Youthed and Moran 1969; Botz et al. 2003; Author's personal copy

ARTICLE IN PRESS 6 B. Mencinger-Vracˇko, D. Devetak / Zoology 111 (2008) 2–8

Table 1. Circular statistics of sand-tossing behaviour of Euroleon nostras in the presence of prey

Prey species Circular 95% conﬁdence Circular Circular Correlation mean interval (7)of variance standard between (a) direction circular mean deviation and (b) (m) direction

Mealworm Tenebrio Prey angle (a) 182.31 175.51 0.42 601 r ¼ 0.79 molitor – unimpeded 189.11 Po0.0010 N ¼ 91, n ¼ 337 Sand tossing angle (b) 178.71 174.11 0.25 431 183.21 Watson’s F-test for circular F ¼ 0.76; P ¼ 0.3800; n.s. mean directions – difference between (a) and (b) Mealworm Tenebrio Prey angle (a) 179.61 163.71 0.53 711 r ¼ 0.80 molitor with covered eyes 195.61 Po0.0010 N ¼ 32, n ¼ 101 Sand tossing angle (b) 177.41 167.81 0.31 491 186.91 Watson’s F-test for circular F ¼ 0.06; P ¼ 0.8000; n.s. mean directions – difference between (a) and (b) Woodlouse Trachelipus Prey angle (a) 196.31 182.91 0.43 611 r ¼ 0.86 rathkei N ¼ 23, n ¼ 87 209.81 Po0.0010 Sand-tossing angle (b) 191.81 181.81 0.29 481 201.91 Watson’s F-test for circular F ¼ 0.28; P ¼ 0.6000; n.s. mean directions – difference between (a) and (b) Firebug Pyrrhocoris Prey angle (a) 179.61 163.71 0.53 711 r ¼ 0.78 apterus N ¼ 10, n ¼ 28 195.61 Po0.0010 Sand-tossing angle (b) 177.41 167.81 0.31 491 186.91 Watson’s F-test for circular F ¼ 0.13; P ¼ 0.7200; n.s. mean directions – difference between (a) and (b) Ant Formica sp. N ¼ 11, Prey angle (a) 226.41 184.71 0.60 781 r ¼ 0.88 n ¼ 21 268.11 Po0.0010 Sand-tossing angle (b) 198.41 168.61 0.47 641 228.21 Watson’s F-test for circular F ¼ 1.37; P ¼ 0.2500; n.s. mean directions – difference between (a) and (b)

Abbreviations: N ¼ number of antlions, n ¼ number of trials; n.s. ¼ not signiﬁcant.

Matsura et al. 2005; Devetak et al. 2005, 2007a). In fine locomotion is impeded . Therefore, antlions prefer sands sand, vibrations are more strongly attenuated than in of medium particle size (Botz et al. 2003; Matsura et al. coarse sand so that an antlion detects its prey only at a 2005; Devetak et al. 2005) and that was the reason why shorter distance (Devetak et al. 2007b). Ants escape this sand was used in the present study. faster from pits dug in coarse-grained sand than from The presented results of the blinding experiments pits dug in fine-grained sand (Botz et al. 2003; Farji- make it very clear that antlions do not use visual cues for Brener 2003). Substrate particle size has a significant effect their sand-tossing behaviour. Antlions have larval eyes, on the angle of the pit wall (Allen and Croft 1985; Botz et with stemmata on their eye tubercles (Jockusch 1967). al. 2003); a steeper wall retards the escape of ants. Taking Even after the eyes of the antlions were covered with all of these factors into account, antlions compromise in paint to exclude visual stimuli the larvae still responded their choice of sand . Despite the longer reaction distance, with normal predatory behaviour to the presence of prey coarser sand is inconvenient as potential prey escapes and located it without seeing it. The use of olfactory more easily (Lucas 1982; Botz et al. 2003)andlarval cues may also be excluded since they would not allow Author's personal copy

ARTICLE IN PRESS B. Mencinger-Vracˇko, D. Devetak / Zoology 111 (2008) 2–8 7 determination of the position and the distance of the until after sand tossing is completed. To face its prey, prey as accurately as found in the present experiments. the antlion in some cases needs to turn 1801. Further Thus, the only cue the antlion could have used is studies are needed to elucidate how the antlion is able to substrate vibration. Despite the clear behavioural cope with the problem that the prey is behind it. It evidence, organs acting as receptors for substrate would be interesting to study the behaviour of prey and vibration could not yet be localised. However, it is most predator immediately after sand tossing. likely that the sensory hair organs on the mesothorax and metathorax are directly involved in vibration detection (Le Faucheux 1972). The present results show for the first time that Acknowledgements locomotory activity of prey outside the pit evokes aimed We are grateful to Prof. Dr. Karl Kral (University of sand-tossing behaviour of antlions. The antlion larvae Graz) for helpful discussions and to Dr. Michael Poteser oriented very precisely toward the source of vibratory (Graz) for assistance with the computer program Mirror signals since the correlation between the target angles Video-Analysis for Windows. We thank Dr. T. Matsura and the sand-tossing angles was high. The small reaction and an anonymous reviewer for their valuable sugges- distances for sand tossing, as found in this study, may be tions and comments on a preliminary version of explained by the attenuation values of signals in the manuscript. This study was partly funded by a dry sand. Propagation velocities of Rayleigh-waves research grant from the Slovene Ministry of High (R-waves) travelling in the sand used in this study Education, Science and Technology (Grant no. P1- amount to 25–35 m/s and depend on the frequency of 0078 Biodiversity). the signals (Devetak, unpublished data). Reaction distances differ significantly for different prey and there is a positive correlation between prey size and distance. The mechanism used by antlion larvae to determine the References prey angle has not yet been specifically studied. Due to the low propagation velocities of R-waves the time and Aicher, B., Tautz, J., 1990. Vibrational communication in the phase differences of the signals at the receptors on both fiddler crab, Uca pugilator. I. Signal transmission through sides of the body may be expected to determine the prey the substratum. J. Comp. Physiol. A 166, 345–353. Allen, G.R., Croft, D.B., 1985. Soil particle size and the pit angle. Additionally, differences in signal amplitudes morphology of the Australian antlions Myrmeleon dimin- could contribute to the accuracy of prey localization. utus and M. pictifrons (Neuroptera: Myrmeleontidae). For sands inhabited by the antlion Euroleon it was Austr. J. Zool. 33, 863–874. found that the attenuation of signals with biologically Barth, F.G., 1986. Vibrationssinn und vibratorische Umwelt relevant frequencies (around 300 Hz) varies from 0.26 to von Spinnen. Naturwissenschaften 73, 519–530. 2.6 dB/cm and is inversely proportional to sand particle Batschelet, E., 1981. Circular Statistics in Biology. Academic size (Devetak et al. 2007b). Attenuation values of similar Press, New York. magnitude were recorded in the vibratory communica- Bongers, J., Koch, M., 1981. Trichterbau des Ameisenlo¨wen tion of spiders (on banana leaves; Barth 1986), scorpions Euroleon nostras Fourcr. Neth. J. Zool. 31, 329–341. (in dry sand; Brownell 1977), and fiddler crabs (in wet Botz, J.T., Loudon, C., Barger, J.B., Olafsen, J.S., Steeples, sand; Aicher and Tautz 1990). D.W., 2003. Effects of slope and particle size on ant locomotion: implications for choice of substrate by At the present it is not possible to determine the antlions. J. Kansas Entomol. Soc. 76, 426–435. functional significance of sand tossing towards prey Brownell, P.H., 1977. Compressional and surface waves in moving outside the pit. It is likely that prey close to the sand: used by desert scorpions to locate prey. Science 197, pit is disoriented by sand bombardment or even directed 479–482. towards the pit. Additional studies, however, are needed Brownell, P., Farley, R.D., 1979a. Prey localizing behaviour of to establish this. the nocturnal desert scorpion, Paruroctonus mesaensis: The ability of the larva to toss sand results from the orientation to substrate vibrations. Anim. Behav. 27, flexibility of the intersegmental membranes of the thorax 185–193. (Bongers and Koch 1981). The antlion larva is able to Brownell, P., Farley, R.D., 1979b. Orientation to vibrations in toss sand in the posterior direction but unable to toss it sand by nocturnal scorpion Paruroctonus mesaensis: me- anteriorly. This asymmetry in response is explained by chanism of target localization. J. Comp. Physiol. 131, 31–38. the movement mechanics of the head, mandibles and the Devetak, D., 1985. Detection of substrate vibrations in the prothorax as these parts of the body serve as a shovel. antlion larva, Myrmeleon formicarius (Neuroptera: Myrme- While the front part of the antlion is easily moved leonidae). Biol. Vestn. 33, 11–22. dorsoventrally enabling sand tossing in the posterior Devetak, D., 2005. Effects of larval antlions Euroleon nostras direction, sand tossing is not possible in the anterior (Neuroptera, Myrmeleontidae) and their pits on the escape direction. The antlion larva cannot turn toward the prey time of ants. Physiol. Entomol. 30, 82–86. Author's personal copy

ARTICLE IN PRESS 8 B. Mencinger-Vracˇko, D. Devetak / Zoology 111 (2008) 2–8

Devetak, D., Sˇpernjak, A., Janzˇekovicˇ, F., 2005. Substrate Le Faucheux, M., 1972. Le role des soies thoraciques dans la particle size affects pit building decision and pit size in the capture des proies par la larve d’Euroleon nostras Fourcroy antlion larvae Euroleon nostras (Neuroptera: Myrmeleonti- (Ne´vropte` re). Rev. Comp. Anim. 6, 217–221. dae). Physiol. Entomol. 30, 158–163. Lucas, J.R., 1982. The biophysics of pit construction by Devetak, D., Mencinger-Vracˇko, B., Sˇpernjak, A., Devetak, antlion larvae (Myrmeleon, Neuroptera). Anim. Behav. 30, M., 2007a. Capture success in pit-building antlion Euroleon 651–664. nostras (Geoffroy in Fourcroy, 1785) (Neuroptera: Myr- Mansell, M.W., 1996. Predation strategies and evolution in meleontidae) depends on the presence of pits, sand particle antlions (Insecta: Neuroptera: Myrmeleontidae). In: Ca- size and transmission of vibratory signals: a mini-review. nard, M., Aspo¨ck, H., Mansell, M.W. (Eds.), Pure and Ann. Mus. Civ. Storia Nat. Ferrara 8, 161–165. Applied Research in Neuropterology. Sacco, Toulouse, Devetak, D., Mencinger-Vracˇko, B., Devetak, M., Marhl, M., pp. 161–169. Sˇpernjak, A., 2007b. Sand as a medium for transmission of Mansell, M.W., 1999. Evolution and success of antlions vibratory signals of prey in antlions (Myrmeleontidae). (Neuropterida: Neuroptera: Myrmeleontidae). Stapfia 60, Physiol. Entomol. 32, 268–274. 49–58. Farji-Brener, A.G., 2003. Microhabitat selection by antlion Matsura, T., Yamaga, Y., Itoh, M., 2005. Substrate selection larvae, Myrmeleon crudelis: effect of soil particle size on pit- for pit making and oviposition in an antlion, Myrmeleon trap design and prey capture. J. Insect Behav. 16, 783–796. bore Tjeder, in terms of sand particle size. Entomol. Sci. 8, Fertin, A., Casas, J., 2006. Efficiency of antlion trap 347–353. construction. J. Exp. Biol. 209, 3510–3515. Mencinger, B., 1998. Prey recognition in larvae of the antlion Fisher, N.I., 1995. Statistical Analysis of Circular Data. Euroleon nostras (Neuroptera: Myrmeleontidae). Acta Cambridge University Press, Cambridge. Zool. Fenn. 209, 157–161. Gepp, J., Ho¨lzel, H., 1989. Ameisenlo¨wen und Ameisenjung- Scharf, I., Ovadia, O., 2006. Factors influencing site abandon- fern – Myrmeleonidae. Ziemsen Verlag, Wittenberg. ment and site selection in a sit-and-wait predator: a review Griffiths, D., 1980. The feeding biology of antlion larvae: prey of pit-building antlion larvae. J. Insect Behav. 19, 197–218. capture, handling and utilization. J. Anim. Ecol. 49, Sokal, R., Rohlf, F.J., 1995. Biometry: The Principles and 99–125. Practice of Statistics in Biological Research. W.H. Freeman Jockusch, B., 1967. Bau und Funktion eines larvalen and Company, New York. Insektenauges. Untersuchungen am Ameisenlo¨wen (Euro- Topoff, H., 1977. The pit and the antlion. Nat. Hist. 86, 64–71. leon nostras Fourcroy, Planip., Myrmel.). Z. Vergl. Physiol. Youthed, G.J., Moran, V.C., 1969. Pit construction by 56, 171–198. myrmeleontid larvae. J. Insect Physiol. 15, 867–875. This article was published in an Elsevier journal. The attached copy is furnished to the author for non-commercial research and education use, including for instruction at the author’s institution, sharing with colleagues and providing to institution administration. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: