AMER.ZOOL., 41:1215±1221 (2001)

Vibrational Communication and the Ecology of Group-Living, Herbivorous Insects1

REGINALD B. COCROFT2 Division of Biological Sciences, 105 Tucker Hall, University of Missouri, Columbia, Missouri 65211

SYNOPSIS. Communication among members of a colony is a key feature of the success of eusocial insects. The same may be true in other forms of insect sociality. I suggest that substrate-borne vibrational communication is important in the suc- cess of group-living, herbivorous insects. I examine three challenges encountered by herbivorous insects: locating and remaining in a group of conspeci®cs; locating food resources; and avoiding predation. Studies of groups of immature treehop- pers, saw¯ies and butter¯ies suggest that vibrational communication can be im- portant in each of these contexts, enhancing the ability of these group-living her- bivores to exploit the resources of their host plants.

INTRODUCTION cause there often are considerable bene®ts The ecological importance of eusocial in- to individuals of living in groups, one chal- sects such as bees, ants and termites is due lenge is to locate and remain with other in- in part to their remarkable ability to monitor dividuals. Second, because the location of changing resources in their environment high-quality feeding sites will vary over (HoÈlldobler and Wilson, 1990; Seeley, time within a host plant, another challenge 1995; Shellman-Reeve, 1997). The ability is to locate currently pro®table feeding of an insect colony to ef®ciently exploit un- sites. Finally, herbivorous insects must predictable resources is, in turn, based on avoid predation. I will suggest that, in many elaborate systems of communication among species, vibrational communication among colony members. Accordingly, one of the group members is important for solving hallmarks of the eusocial insects is that col- each of these challenges. ony members communicate in relation to important features of their environment BENEFITS OF GROUP LIVING (Seeley, 1995). The eusocial insects, how- Although there are inherent disadvantag- ever, represent only one end of a broad es to group living, such as increased com- spectrum of insect sociality. Analogous petition and risk of disease (Alexander, communication systems can exist in very 1974), plant-feeding insects may bene®t in different forms of insect society, as shown, various ways from being in a group. Pro- for example, by studies of trail-marking tection against predators has been proposed pheromones in group-living lepidopteran to be one of the most general factors se- and saw¯y larvae (Fitzgerald, 1995; Costa lecting for group living (Hamilton, 1971; and Louque, 2001). Here I will suggest that Alexander, 1974; Vulinec, 1990; Mooring for some (and perhaps many) group-living and Hart, 1992). In insects, this might oc- insects that feed on plants, substrate-borne cur, for example, through dilution effects vibrational communication is an important (Foster and Treherne, 1981) or through en- component of their ability to exploit host hancement of chemical defenses (e.g., Ad- plant resources. rich and Blum, 1978). For some herbivo- I focus on three challenges faced by rous insects, feeding ef®ciency is increased group-living, herbivorous insects. First, be- by the presence of conspeci®cs (Ghent, 1960; Kalin and Knerer, 1977; Lawrence, 1 From the Symposium Vibration as a Communi- 1990), resulting in faster growth rates and/ cation Channel presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 3±7 or greater survivorship. Other bene®ts of January 2001, at Chicago, Illinois. grouping can include increased water up- 2 E-mail: [email protected] take (Lockwood and Story, 1986), slower 1215 1216 REGINALD B. COCROFT water loss (Friedlander, 1965), and en- Australian saw¯y Perga dorsalis, larvae hanced thermoregulation (Seymour, 1974). (sometimes called ``spit®res'') form groups Indeed, Costa and Pierce (1997) suggest that move not only within a single tree, but that there may often be suf®cient direct also from one tree to another. According to bene®ts of group living that grouping is fa- Carne (1962), individual P. dorsalis larvae vored whether or not the individuals are ge- in migrating groups continually assess the netically related. One line of evidence sup- presence of nearby individuals by ``tap- porting this view is that, in many species, ping'' with a hardened sclerite at the end of groups that encounter each other merge into their abdomen: ``If an individual strays a larger group composed of individuals from the moving column and fails to make from different family groups, species, or contact with another larva, it manifests dis- genera (Carne, 1962; Wood, 1984, 1993). turbance by an abrupt increase in its rate of tapping. The larvae in the main body of the LOCATING AND REMAINING IN A GROUP colony respond immediately by uncoordi- In some cases, the individuals on the nated tapping for a period of 10±15 sec. same plant may be in groups from the start, There is usually an ``answering'' signal if they hatch from eggs laid in a cluster. from the stray, then further tapping on the However, in other cases, groups are com- part of the colony. It seems certain that this posed of individuals hatching from eggs de- is a form of communication for it invariably posited in different locations (e.g., the tree- results in the individual rejoining its colo- hopper Vanduzea arquata; [Fritz, 1982]). ny.'' Once the individual rejoins the colony, Furthermore, groups may move from one tapping activity subsides. Carne (1962) fur- location to another (e.g., Carne, 1962). ther suggests larvae respond not to the air- Consequently, individuals will often be borne sound, but to the vibration produced faced with the challenge of locating or re- by tapping. Evans (1934) suggested that joining a group. Several lines of evidence tapping occurs in a similar context during suggest that this task can be accomplished group movements in other species in the ge- by means of vibrational communication. nus Perga. First, is it possible for a small insect to A strikingly similar pattern has been ob- detect the location of a vibration source? In served in the chrysomelid beetle Polychal- many cases, the answer is yes. There is ex- ma multicava (D. Windsor, personal com- tensive evidence that insects can locate a munication). In this species, groups of lar- vibration source to one of two stems at a vae migrate from resting positions at the branching point (Latimer and Schatral, base of small plants to feeding areas at the 1983; Steidl and Kalmring, 1989; Ota and tips. When individuals become separated at Cokl, 1991; Roces et al., 1993; Pfannenstiel a branching point, the two groups re-aggre- et al., 1995). This ability is not surprising, gate after back-and-forth bouts of substrate given the large number of taxa in which tapping. males localize receptive females by means Vibrational signaling during group of plant-borne vibrations (Michelsen et al., movements may occur in the tingid bug 1982; Markl, 1983; Claridge, 1985; Gogala, Corythucha hewitti, in which groups of 1985; Henry, 1994; Stewart, 1997). There nymphs are attended by a female. Faeth is also indirect evidence that some insects (1989) observed that disturbance of the leaf can determine whether a vibration source is containing an aggregation of C. hewitii in front them or behind them on a single, caused a nymph to stop feeding and move unbranched stem (Cokl et al., 1999; see dis- away, ``occasionally stopping and vibrating cussion in Cocroft et al., 2000). its abdomen in the vertical plane. Other What evidence is there that insects use nymphs in the brood followed.'' Because plant-borne vibrational cues to locate a such abdominal vibrations are involved in group of conspeci®cs? Observations sug- signal production in other insects (e.g., gest that group-living saw¯y larvae use vi- Henry, 1994), and because such movements brational signals to rejoin a moving group will unavoidably produce a vibration in the from which they become separated. In the substrate, these observations suggest the COMMUNICATION IN HERBIVOROUS INSECTS 1217

FIG. 1. Audiospectrograms of plant-borne vibrational signals. (A) A signal produced by a nymph of the tree- hopper Calloconophora pinguis after having located a FIG. 2. An aggregation of nymphs of the treehopper high-quality feeding site; (B) A coordinated, group sig- Umbonia crassicornis on a host plant stem. nal from an aggregation of nymphs of the treehopper Umbonia crassicornis, produced in response to the ap- proach of a predator; (C) A series of signals produced appears to allow sibling groups to take ad- by an (unidenti®ed) ant-attended lycaenid caterpillar. vantage of changing nutritional resources on the plant. It also shows that locating a feeding site can, in some circumstances, be production of vibrational signals in the con- essentially the same task as locating a text of group movement. group. The only additional requirement for food recruitment is that group-location sig- LOCATING A FOOD RESOURCE nals are produced at an appropriate feeding In the membracid treehoppers Callocon- site. ophora caliginosa and C. pinguis, nymphs Hograefe (1984) reported that larvae of develop to adulthood in tight aggregations, the saw¯y Hemichroa crocea, which live in accompanied at least in the early nymphal groups on birch and alder, communicate stages by their mother (Wood, 1978 [as Gu- while feeding. A signal is produced as the ayaquila compressa]; R.B.C., unpublished end of the abdomen is repeatedly scraped data). These treehopper groups have a no- against the leaf surface in a characteristic madic foraging pattern, in which the entire rhythmic pattern. Signaling is more fre- group moves from one feeding site to an- quent when larvae are on new, undamaged other. In C. pinguis, aggregated nymphs on leaves, which represent high-quality feed- a stem whose nutritional quality is declin- ing sites, and less frequent when larvae are ing (such as a maturing stem or a cut stem) on already heavily damaged leaves. Larvae eventually leave the group and explore the eventually move from low to high quality rest of the plant, probing with their mouth- sites, apparently orienting by means of the parts. When a nymph encounters a suitable vibrational signals. Again, orientation to a feeding site (a new, growing shoot), it stops group and to a feeding site are closely re- and produces a long series of vibrational lated. signals (see Fig. 1A). These signals are used by other individuals to locate the new DEFENSE AGAINST PREDATORS:PARENT- feeding site (R.B.C., unpublished data). In- OFFSPRING INTERACTIONS dividuals in the dispersing group alternate In the membracid treehopper Umbonia periods of walking toward the source with crassicornis, females defend their nymphal periods of quiescence: they wait until a sig- offspring from predators. Nymphs develop nal is perceived, then walk, then wait for to maturity in a dense aggregation of up another signal. Individuals that arrive at the to100 individuals encircling a host plant site begin to signal in unison with the in- stem (Fig. 2). In their exposed position near dividuals already there. The alternation of the tip of a growing shoot, the nymphs are walking and waiting to detect a signal was preyed upon by a diverse array of inverte- also observed in saw¯y larvae locating a brates such as predatory Hemiptera, syrphid group (Carne, 1962). This signaling system ¯y larvae, coccinellid beetle larvae, spiders, 1218 REGINALD B. COCROFT and wasps (Wood, 1974, 1976, 1983; Dow- have important consequences for membra- ell and Johnson, 1986; McKamey and cid social behavior: it is correlated with, Deitz, 1996; Cocroft, 2002). Females de- and enhanced by, aggregating as opposed fend their offspring by approaching the to solitary behavior (McEvoy, 1979). predator from their position outside the Mutualism with ants, then, will often se- group, fanning their wings, and kicking lect for aggregating behavior by treehop- with their hind legs (Wood, 1976, 1983; pers. In the Neotropics, groups often consist Dowell and Johnson, 1986; Cocroft, of more than one species (Wood, 1984), 1999b). Maternal defense is an important again highlighting the likely importance of resource for offspring: the female is their direct bene®ts to the grouped individuals. only protection against invertebrate preda- We might expect, then, that ant-attended tors, and if she disappears the nymphs' species will have signals used in the for- chances of survival are low (Wood, 1976; mation and maintenance of groups. Such Dowell and Johnson, 1986; Cocroft, 2002). signals appear to be present in at least some When a predator approaches, nymphs of membracid species that form mutualisms U. crassicornis produce vibrational signals with ants (e.g., species in several genera in (Cocroft, 1996, 1999a, b). The signal of the subfamily Membracinae; R.B.C., un- one nymph is a brief series of pulses, last- published data). ing about 50 msec. However, the nymphs Vibrational communication is an impor- within an aggregation coordinate their sig- tant component of the mutualism of lycaen- nals (Fig. 1B). Signaling usually begins on id and riodinid butter¯y larvae with ants one end of the aggregation (e.g., where the (Fig. 1C; deVries, 1990; Travassos and predator ®rst contacts a nymph) and travels Pierce, 2000). The signals of larval and pu- across the group in a rapid wave as indi- pal lycaenids appear to function in attract- viduals respond to the signals of their ing and maintaining an association with neighbors. As a result, signals of individu- ants (deVries, 1991; Travassos and Pierce, als merge into a characteristic group display 2000). Travassos and Pierce (2000) also that is longer and higher in amplitude than tested the hypothesis that signals were in- the signal of a single nymph (Cocroft, volved in the formation of groups; however, 1999a). Females respond to these coordi- their results did not suggest a direct role of nated signals by quickly moving into the the signals in attracting conspeci®cs. As in aggregation. In one ®eld study, the combi- these lepidopterans, signaling to attract mu- nation of signaling nymphs and defending tualistic ants might also be expected in females was successful in repelling about 3 membracids. Whether membracids signal to out of 4 attacks by predatory wasps (Co- attract ants is unknown, although observa- croft, 2002). The coordination of signals tions suggest that this may be the case for among nymphs is necessary to elicit the fe- nymphs of one ant-attended Neotropical male's response (Cocroft, 1996). Because species (Tomogonia vittatipennis; R.B.C., obtaining the bene®ts of maternal defense unpublished data). requires collective effort, this signaling be- havior contains an element of cooperation. GENERAL DISCUSSION One common feature of many of the DEFENSE AGAINST PREDATORS:ANT communication systems described here (es- MUTUALISM pecially in saw¯y larvae, membracids, and Many species of membracid treehoppers, chrysomelid beetle larvae) is the simulta- especially in the tropics, have mutualistic neous production of signals by multiple in- relationships with honeydew-harvesting dividuals. This process also occurs in the ants (Wood, 1984, 1993). This relationship chemical signals produced during recruit- can have important consequences for mem- ment communication in some eusocial in- bracid survival, because at least some ant sects and in tent caterpillars (Costa and species greatly reduce treehopper mortality Pierce, 1997; HoÈlldobler and Wilson, from other predatory insects (McEvoy, 1990). This coordination of signaling may 1979; Fritz, 1982). Ant-mutualism may provide a means by which individuals with COMMUNICATION IN HERBIVOROUS INSECTS 1219 a common interest can enhance the signals the joint signaling of these nymphs may in- of other colony or group members (see Cos- deed represent a case of signal enhance- ta and Pierce, 1997). However, although ment. Further examination of the costs and group members will often have an overlap bene®ts of group living, and of the dynam- of reproductive interests, this will not al- ics of signaling, will be needed to resolve ways be true. Although there may be ben- the issue. In general, resolving the interplay e®ts of group living, these bene®ts will of- of cooperation and con¯ict in the signaling ten be unequally shared among group mem- interactions of group members will proba- bers (reviewed in Krause, 1994). The bly require a case-by-case examination. nymphal aggregations of U. crassicornis In many insect groups, chemical com- treehoppers provide an illustration. Al- munication plays a role similar to that of though maternal defense is important in re- the vibrational communication systems de- ducing predation, the distribution of pre- scribed here. In some group-living insect dation risk in offspring aggregations is far herbivores, chemical cues attract individu- from uniform. Individuals on the edges of als to groups of conspeci®cs (Aldrich and aggregations, and those farther from the fe- Blum, 1978). In group-living lepidopteran male at the time they are contacted by a and saw¯y larvae, a complex system of predator, are substantially more likely to be chemical trail-marking underlies their for- preyed on than individuals in the center of aging behavior (Fitzgerald, 1995; Costa and the aggregation and/or closer to the female Louque, 2001). Chemical cues are often im- (Cocroft, 2002). While there is a clear ele- portant in anti-predator defense in group- ment of cooperation in offspring signaling, living species, in which alarm pheromones then, the ®nding that predation risk is un- and/or cues associated with injury alert oth- equally distributed suggests that coopera- er group members to the presence of a pred- tion may have its limits. This would be es- ator (Nault and Phelan, 1984). In some in- pecially true if offspring are able to in¯u- sects with parental care (e.g., the treehopper ence the female's position during an attack. Umbonia crassicornis), both chemical cues Although it is currently not known whether (Wood, 1976) and vibrational signals (Co- competition for access to maternal defense croft, 1999a) can elicit parental defense of is even possible in this species, it is clear offspring, and there is likely to be an inter- that there is at least the potential for con¯ict action between the two kinds of signals in among group members. A divergence of in- their effect on parental responses to preda- terests may also occur in other groups, such tors. as those of migrating saw¯y larvae. For ex- Although evidence for the role of vibra- ample, if edge individuals are more vulner- tional communication is anecdotal or lack- able to predators or parasitoids, individuals ing for many group-living herbivorous in- in that position may be more likely to signal sects, studies of membracids, saw¯ies, and to attract `strays' than individuals in the lepidopteran larvae suggest that this form center. Indeed, if strays will become edge of communication may represent an impor- individuals, shielding those currently on the tant set of adaptations to herbivory. Anal- edges, it is conceivable that edge individu- ogous communication systems may be pre- als could compete among each other to at- sent in other groups, many of which are tract strays to their location. Carne (1962) known to use vibrational signals in com- reports that after strays had rejoined a mi- munication in at least some contexts. Par- grating group, the only individuals that still ent-offspring communication in response to signaled were those on the edges of the predators might be especially likely in so- group. cial Hemiptera, chrysomelid beetles, and In other cases, such as in the food re- saw¯ies with maternal care (Dias, 1975, cruitment signals of nymphs of the treehop- 1976; Windsor, 1987; Kudo, 1990; Kudo et per Calloconophora pinguis, it may be in al., 1995; Tallamy and Schaeffer, 1997). Vi- the interests of all of the individuals re- brational communication among group cruited to a food source to recruit the re- members might also be expected in taxa maining individuals in the group. If so, then such as many lepidopterans (Costa and 1220 REGINALD B. COCROFT

Pierce, 1997), some aphids (Williams, Costa, J. T. and R. W. Louque. 2001. Group foraging 1922; Eastop, 1954), Neuroptera (Henry, and trail following behavior of the red-headed pine saw¯y Neodiprion leconei (Fitch) (Hymen- 1972), tingid bugs (Faeth, 1989), and ant- toptera: Symphyta: Diprionidae). Ann. Entomol. attended cicadellids (Dietrich and Mc- Soc. Am. 94:480±489. Kamey, 1990) and fulgoroids (Bourgoin, Costa, J. T. and N. E. Pierce. 1997. Social evolution 1997). Only further study of communica- in the Lepidoptera: Ecological context and com- munication in larval societies. In J. C. Choe and tion in these fascinating insect societies will B. J. Crespi (eds.), The evolution of social behav- reveal the extent to which vibrational com- ior in insects and arachnids, pp. 407±442. Cam- munication is a widespread adaptation to bridge University Press, Cambridge. the challenges of herbivory in group-living deVries, P. J. 1990. Enhancement of symbioses be- tween butter¯y caterpillars and ants by vibrational insects. communication. Science 248:1104±1106. deVries, P. J. 1991. Mutualism between Thisbe irenea ACKNOWLEDGMENTS butter¯ies and ants and the role of ant ecology in I thank Peggy Hill for the opportunity to the evolution of larval-ant associations. Biol. J. participate in this symposium. The manu- Linnean Soc. 43:179±196. Dias, B. F. de Souza. 1975. Comportamento pre-social script bene®ted from the comments of J. de Sin®tas do Brasil Central. I. Themos olfersii Edwards, two anonymous reviewers, and (Klug) (Hymenoptera, Argidae). Studia Ent. 18: from discussions with J. Yack, R. Hunt, P. 401±432. Hill, and other symposium participants. Dias, B. F. de Souza. 1976. Comportamento pre-social de Sin®tas do Brasil Central. II. Dielocerus diasi Smith, 1975 (Hymenoptera, Argidae). Studia Ent. REFERENCES 19:461±501. Aldrich, J. R. and M. S. Blum. 1978. Aposematic ag- Dietrich, C. H. and S. H. McKamey. 1990. Three new gregation of a bug (Hemiptera: Coreidae): The de- idiocerine leafhoppers (Homoptera: Cicadellidae) fensive display and formation of aggregations. from Guyana with notes on ant-mutualism and Biotropica 10:58±61. subsociality. Proc. Entomol. Soc. Wash. 92:214± Alexander, R. D. 1974. The evolution of social behav- 223. ior. Annu. Rev. Ecol. Syst. 5:325±383. Dowell, R. V. and M. Johnson. 1986. Polistes major Burgoin, T. 1997. Habitat and ant-attendance in He- (Hymenoptera: Vespidae) predation of the tree- miptera: A phylogenetic, test with emphasis on hopper, Umbonia crassicornis (Homoptera: Mem- trophobiosis in Fulgoromorpha. In P. Grandcolas bracidae). Pan-Paci®c Entomol. 62:150±152. (ed.), The origin of biodiversity in insects: Phy- Eastop, V. F. 1954. A sound production mechanism in logenetic tests of evolutionary scenarios. Mem. the Aphididae and the generic position of the spe- Mus. Natn. Hist. Nat. 173:109±124. cies possessing it. Entomologist 85:57±61. Carne, P. B. 1962. The characteristics and behaviour Evans, J. W. 1934. Notes on the behaviour of larval of the saw¯y Perga af®nis af®nis (Hymenoptera). communities of Perga dorsalis Leach (Hymenop- Australian J. Zool. 10:1±34. tera, Tenthredinidae). Trans. R. Ent. Soc. London 82:455±460. Claridge, M. F. 1985. Acoustic signals in the Homop- Faeth, S. H. 1989. Maternal care in a lace bug, Cor- tera: Behavior, taxonomy, and evolution. Annu. ythucha hewitti (Hemiptera: Tingidae). Psyche 96: Rev. Entomol. 30:297±317. 101±110. Cocroft, R. B. 1996. Insect vibrational defense signals. Fitzgerald, T. D. 1995. The tent caterpillars. Cornell Nature 382:679±680. University Press, Ithaca, New York. Cocroft, R. B. 1999a. Offspring-parent communication Foster, W. A. and J. E. Treherne. 1981. Evidence for in a subsocial treehopper (Hemiptera: Membraci- the dilution effect in the sel®sh herd from ®sh dae: Umbonia crassicornis). Behaviour 136:1±21. predation on a marine insect. Nature 295:466± Cocroft, R. B. 1999b. Parent-offspring communication 467. in response to predators in a subsocial treehopper Friedlander, C. P. 1965 Aggregation in Oniscus asellus. (Hemiptera: Membracidae: Umbonia crassicor- Anim. Behav. 13:342±346. nis). Ethology 105:553±568. Fritz, R. S. 1982. An ant-treehopper mutualism: Ef- Cocroft, R. B. 2002. Antipredator defense as a limited fects of Formica subsericea on the survival of resource: Unequal predation risk in broods of an Vanduzea arquata. Ecol. Entomol. 7:267±276. insect with maternal care. Behav. Ecol. (In press) Ghent, A. W. 1960. A study of the group-feeding be- Cocroft, R. B., T. D. Tieu, R. R. Hoy, and R. N. Miles. havior of larvae of the jack pine saw¯y, Neodi- 2000. Directionality in the mechanical response to prion pratti banksianae Roh. Behaviour 16:110± substrate vibration in a treehopper (Hemiptera: 148. Membracidae: Umbonia crassicornis). J. Comp. Gogala, M. 1985. Vibrational communication in in- Physiol. 186:695±705. sects (biophysical and behavioural aspects). In K. Cokl, A., M. Virant-Doberlet, and A. McDowell. 1999. Kalmring and N. Elsner (eds.), Acoustic and vi- Vibrational directionality in the southern green brational communication in insects, pp. 117±126. stink bug, Nezara viridula (L.), is mediated by fe- Proceedings XVII International Congress of En- male song. Anim. Behav. 58:1277±1283. tomology. Parey, Berlin. COMMUNICATION IN HERBIVOROUS INSECTS 1221

Hamilton, W. D. 1971. Geometry for the sel®sh herd. Pfannenstiel, R. S., R. E. Hunt, and K. V. Yeargan. J. Theor. Biol. 31:295±311. 1995. Orientation of a hemipteran predator to vi- Henry, C. S. 1972. Eggs and repagula of Ululodes and brations produced by feeding caterpillars. J. Insect Ascaloptynx (Neuroptera: Ascalaphidae): A com- Behav. 8:1±9. parative study. Psyche 79:1±22. Roces, F., J. Tautz, and B. HoÈlldobler. 1993. Stridula- Henry, C. S. 1994. Singing and cryptic speciation in tion in leaf-cutting ants: Short-range recruitment insects. Trends Ecol. Evol. 9:388±392. through plant-borne vibrations. Naturwissenschaf- Hoegraefe, T. 1984. Substratum-stridulation in the co- ten 80:521±524. lonial saw¯y larvae of Hemichroa crocea (Hy- Seeley, T. D. 1995. The wisdom of the hive: The social menoptera, Tenthredinidae). Zoologischer Anzeig- physiology of honey bee colonies. Harvard Uni- er 213:234±241. versity Press, Cambridge, Massachusetts. HoÈlldobler, B. and E. O. Wilson. 1990. The ants. Har- Seymour, R. S. 1974. Convective and evaporative vard University Press, Cambridge, Massachusetts. cooling in saw¯y larvae. J. Insect Physiol. 20: Kalin, M. and G. Knerer. 1977. Group and mass effects 2447±2457. in diprionid saw¯ies. Nature 267:427±429. Shellman-Reeve, J. S. 1997. The spectrum of euso- Krause, J. 1994. Differential ®tness returns in relation ciality in termites. In J. C. Choe and B. J. Crespi to spatial position in groups. Biol. Rev. 69:187± (eds.), The evolution of social behavior in insects 206. and arachnids, pp. 52±93. Cambridge University Kudo, S. 1990. Brooding behavior in Elasmucha pu- Press, Cambridge. toni (Heteroptera: Acanthosomatidae), and a pos- Steidl, O. and K. Kalmring. 1983. The importance of sible nymphal alarm substance triggering guarding song and vibratory signals in the behaviour of the bushcricket Ephippiger ephippiger Fiebig (Or- responses. Appl. Ent. Zool. 25:431±437. thoptera, Tettigoniidae): Taxis by females. Oec- Kudo, S., E. Ishibashi, and S. Makino. 1995. Repro- ologia 80:142±144. ductive and subsocial behaviour in the ovovivip- Stewart, K. W. 1997. Vibrational communication in in- arous leaf beetle Gonioctena sibirica (Coleoptera: sects: Epitome in the language of stone¯ies? Chrysomelidae). Ecol. Entomol. 20:367±373. Amer. Entomol. Summer 1997:81±91. Latimer, W. and A. Schatral. 1983. The acoustic be- Tallamy, D. W. and C. Schaeffer. 1997. Maternal care haviour of the bushcricket Tettigonia cantans. I. in the Hemiptera: Ancestry, alternatives, and cur- Behavioural responses to sound and vibration. Be- rent adaptive value. In J. C. Choe and B. J. Crespi hav. Proc. 8:113±124. (eds.), The evolution of social behavior in insects Lawrence, W. S. 1990. The effects of group size and and arachnids, pp. 94±115. Cambridge University host species on development and survivorship of Press, Cambridge. a gregarious caterpillar Halisidota caryae (Lepi- Travassos, M. A. and N. E. Pierce. 2000. Acoustics, doptera: Arctiidae). Ecol. Entomol. 15:53±62. context and function of vibrational signalling in a Lockwood, J. A. and R. N. Story. 1986. Adaptive lycaenid butter¯y-ant mutualism. Anim. Behav. functions of nymphal aggregation in the southern 60:13±26. green stink bug, Nezara viridula (L.) (Hemiptera: Vulinec, K. 1990. Collective security: Aggregation by Pentatomidae). Environ. Entomol. 15:739±749. insects as a defense. In D. L. Evans and J. O. Markl, H. 1983. Vibrational communication. In F. Schmidt (eds.), Insect defenses: Adaptive mecha- Huber and H. Markl (eds.), Neurobiology and be- nisms and strategies of prey and predators, pp. havioral physiology, pp. 332±353. Springer, Ber- 251±288. State Univ. of New York Press, Albany, lin. New York. McKamey, S. H. and L. L. Deitz. 1996. Generic re- Williams, C. B. 1922. Co-ordinated rhythm in insects; vision of the New World tribe Hoplophorionini with a record of sound production in an aphid. (Hemiptera: Membracidae: Membracinae). Syst. Entomologist 55:173±176. Entomol. 21:295±342. Windsor, D. M. 1987. Natural history of a subsocial McEvoy, P. B. 1979. Advantages and disadvantages to tortoise beetle, Acromis sparsa Boheman (Chry- group living in treehoppers (Homoptera: Membra- somelidae, Cassidinae) in Panama. Psyche 94: cidae). Misc. Publ. Entomol. Soc. Amer. 11:1±13. 127±150. Michelsen, A., F. Fink, M. Gogala, and D. Traue. 1982. Wood, T. K. 1974. Aggregating behavior of Umbonia crassicornis. Can. Entomol. 106:169±173. Plants as transmission channels for insect vibra- Wood, T. K. 1976. Alarm behavior of brooding female tional songs. Behav. Ecol. Sociobiol. 11:169±281. Umbonia crassicornis (Homoptera: Membraci- Mooring, M. S. and B. L. Hart. 1992. Animal grouping dae). Ann. Ent. Soc. Amer. 69:340±344. for protection from parasites: Sel®sh herd and en- Wood, T. K. 1978. Parental care in Guayaquila com- counter-dilution effects. Behaviour 123:173±193. pressa Walker (Homoptera: Membracidae). Psy- Nault, L. R. and P. L. Phelan. 1984. Alarm pheromones che 85:135±145. and sociality in pre-social insects. In W. J. Bell Wood, T. K. 1983. Brooding and aggregating behavior and R. T. Carde (eds.), Chemical ecology of in- of the treehopper, Umbonia crassicornis. Nat. sects, pp. 237±256. Sinauer Associates, Sunder- Geogr. Res. 15:753±758. land, Massachusetts. Wood, T. K. 1984. Life history patterns of tropical Ota, D. and A. Cokl. 1991. Mate location in the south- membracids (Homoptera: Membracidae). Socio- ern greenstink bug, Nezara viridula (Heteroptera: biology 8:299±344. Pentatomidae), mediated through substrate-borne Wood, T. K. 1993. Diversity in the New World Mem- signals on ivy. J. Insect Behav. 4:441±447. bracidae. Annu. Rev. Entomol. 38:409±435.