Behavioral Ecology Vol. 13 No. 2: 154-159
© 2002 International Society for Behavioral Ecology
Specialized predation on plataspid heteropterans in a coccinellid beetle: adaptive behavior and responses of prey attended or not by ants
a Laboratoire d'Ecologie Terrestre (UMR CNRS no. 5552), Université Toulouse III, 118 route de Narbonne, 31062 Toulouse Cedex, France b Laboratoire d'Ethologie et cognition animale (FRE CNRS no. 2382), Université Toulouse III, 118 route de Narbonne, 31062 Toulouse Cedex, France
Address correspondence to A. Dejean. E-mail: dejean{at}cict.fr .
Received 1 January 2001; revised 23 March 2001; accepted 23 March 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Two plataspid hemipteran species proliferated on Bridelia micrantha (Euphorbiaceae). Colonies of Libyaspis sp., never attended by ants, developed on branches, while Caternaultiella rugosa lived at the base of the trunks, mostly in association with Camponotus brutus that attends them in carton shelters. Both plataspid species are prey of the coccinellid beetle Anisolemnia tetrasticta, whose larvae always detected them by contact. When attacked the Libyaspis nymphs cowered, so that the hypertrophied lateral sides of their tergits made contact with the substrate, but the ladybirds slid their long forelegs under these nymphs, lifted them, and bit them on the ventral face. The Caternaultiella nymphs, which do not have hypertrophied extremities of the tergits, tried to escape at contact with the ladybirds, but were rarely successful. To capture them, the ladybirds either adopted the previous behavior or directly grasped then bit them. We noted a graded aggressiveness in the ants toward the ladybirds according to the situation: no aggressiveness on the tree branches; stopping the ladybirds that approached the shelters where the ants attended Caternaultiella; and full attack of ladybirds that tried to capture Caternaultiella nymphs situated outside shelters. The latter behavior can emit an alarm pheromone that triggers the dispersion of their congeners while attracting attending C. brutus workers. Naive workers are not attracted, so we deduce that this behavior is the result of a kind of learning.
Key words: anti-predator behavior, ants, Heteroptera, Coccinellidae, ladybird beetles, learning, predation, trophobiosis.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although many predatory insects are specialized, they can capture a relatively wide spectrum of prey; however, the suitability of the prey varies greatly. "Essential prey" are capable of supporting immature development and adult reproduction, whereas "alternative prey" only enable the predator to survive, with a mixed diet being frequent and profitable (Evans et al., 1999
; Hodek,
1973).
Most ladybird species are predators. The majority of ladybirds are
aphidophagous, while some others are coccidophagous. Both aphids and coccids
generally live in groups, and several species have evolved trophobiotic
relationships with ants (trophobiosis is a symbiotic relationship wherein ants
obtain honeydew from attended insects, and in turn protect them from their
natural enemies; see Hölldobler and
Wilson, 1990). Aphids have developed several defenses against
their predators, including the release of a pheromone that triggers the
dispersal of the group when an individual is attacked. Ladybirds are therefore
confronted with these defenses and in numerous cases with the aggressiveness
of ants attending aphid colonies, but certain ladybird species use a kind of
camouflage to obtain access to ant-attended aphids or even to live as
parasites in ant nests (Corbara et al.,
2001; Dixon, 2000;
Majerus, 1989;
Sloggett and Majerus, 2000;
Sloggett et al., 1998;
Völkl, 1995).
Ladybirds are also predators of plataspid heteropterans. In India the
populations of Coptosoma ostentuum, a pest of cultivated pulses, are
regulated by the ladybird Synia melanaria, and in Australia
Synoma seminigra attacks nymphs of Coptosoma sp.
(Malhotra and Krishnaswami,
1962; Pope, 1988).
Anisolemnia tetrasticta (= Caria shoutedeni) has been cited
as a predator of Libyaspis plataspids on Sesbania sp., a
leguminous shade tree of Uganda
(Hargreaves, 1925). In
Cameroon we found A. tetrasticta preying on Libyaspis sp.
colonies developing on several leguminous tree species and on Bridelia
micrantha (Euphorbiaceae). B. micrantha can also support another
plataspid, Caternaultiella rugosa, which is attended by ants, mostly
Camponotus brutus (Dejean et al.,
2000b). Although most of the known cases of trophobiosis concern
certain lycaenid lepidoptera and Hemiptera of the former suborder Homoptera,
interactions between ants and heteropterans have also been reported, but
mostly through casual observations
(Maschwitz et al., 1987). As a
result, verifying if relationships between ants and certain heteropterans are
reciprocal has become a challenging field of scientific inquiry. In this study
we searched for natural predators of plataspids. We also noted that A.
tetrasticta can prey on C. rugosa attended by ants outside
carton shelters built by C. brutus.
We conducted a two-pronged study concerning firstly the predatory behavior of the ladybird A. tetrasticta when confronted with the two plataspid species and secondly the protective activity of the ants that attend Caternaultiella against the predatory pressure of this ladybird. We attempted to verify whether this ladybird is specialized on one of the plataspids by asking whether the ladybird larvae are able to counter the Libyaspis and/or Caternaultiella defenses, and whether the ladybird larvae trigger reactions by C. brutus workers when attempting to prey on Caternaultiella.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Ant, plataspid, and coccinellid species
Camponotus brutus is a subdominant formicine ant species widely distributed in central Africa. It has been noted in pioneer vegetal formations along forest edges, in the savanna, in cocoa tree plantations, and in the canopies of secondary and old forests. This species, which is nocturnal, can be active both day and night when exploiting large hemipterans (Dejean and Gibernau, 2000
;
Dejean et al., 2000a,
b,
c).
Caternaultiella rugosa is a plataspid heteropteran attended in
shelters built by C. brutus or Myrmicaria opaciventris in
the root area of Bridelia spp. (Euphorbiaceae). During periods of
proliferation, colonies of this C. rugosa develop outside the
shelters. In this case, the females guard their egg masses. Later, adults and
last-instar nymphs place themselves above and around first instars
(Dejean et al., 2000b;
Gibernau and Dejean,
2000).
Libyaspis sp. is a plataspid that develops colonies without associated ants on branches of B. micrantha as well as on Albizia spp. and Cassia spp. (Leguminosae). In this study, colonies of 50-95 individuals generally established themselves on young branches. Adults are well protected by their pronotum and scutellum (see Figure 1). They group at the base of the branches where the colonies develop, so that it is difficult for crawling insects to access the nymphs situated on the distal part of the branches.
|
Anisolemnia tetrasticta subspecies schoutedeni (Coleoptera; Coccinellidae) was found on B. micrantha inhabited by Libyaspis sp. Females laid their eggs on the branches where Libyaspis proliferated. The adults and first-instar larvae fed on the eggs of Libyaspis. After the second moult, larvae preyed only on the first-instar nymphs of Libyaspis, while the larvae of later instars preyed on nymphs of a wide range of sizes. These ladybirds also explored other parts of the trees, thus enabling them to encounter the Caternaultiella at the base of the trunks. Last-instar larvae pupated on the branches of the trees.
Voucher specimens of ants, plataspids, and coccinellids were deposited and identified at the Museum of Natural History, London.
Preparation of experimental conditions
Colonies of Libyaspis sp. observed in different forests of
southern Cameroon, as well as the branches supporting them, were transported
in large plastic bags to the campus of the University Yaoundé I and
Mvolier Valley, Yaoundé, two easily accessible sites where both B.
micrantha and C. brutus were previously studied (see
Dejean et al., 2000b;
Gibernau and Dejean, 2000).
Each branch with a colony of Libyaspis sp. was attached to a branch
of the new supporting tree, onto which the colony spread. The coccinellids
were introduced along with the Libyaspis, then their populations
developed. In this way, we obtained 14 B. micrantha trees sheltering:
(1) C. brutus attending Caternaultiella in shelters; (2)
during periods of proliferation, Caternaultiella clusters developing
at the base of the trunks, outside the shelters; (3) Libyaspis sp.
colonies developing on the upper branches of the same trees; and (4) a
population of A. tetrasticta. The latter developed first on the
branches, preying on Libyaspis, but when the population increased
some last-instar larvae foraged over entire trees and preyed on
Caternaultiella clusters developing outside the shelters.
Predatory behavior of the coccinellid and plataspids' responses
Our studies concentrated on the predatory activity of a total of 193
last-instar larvae of the conccinellid. Ladybird larvae become more voracious
and successful at capturing prey as they grow in size, with fourth-instar
larvae consuming 65% of the total prey required for development
(Dixon, 2000). Ladybirds were
bred in petri dishes (8-cm diam) in which we introduced a 16-cm2
piece of bark from Bridelia. We first conducted a series of 26 choice
tests (26 different ladybirds used) by introducing into the petri dishes both
a Libyaspis and a Caternaultiella nymph of the same size. We
noticed each time the species of the individual attacked first. For the study
of predatory behavior, the prey were introduced one by one. We recorded the
behavioral sequences by direct observation. During preliminary experiments
conducted under natural conditions, we noted the richest behavioral sequences,
permitting us to establish data sheets that we used during experimentation to
record each behavioral act performed (see
Kenne et al., 2000). For each
kind of prey, a flow diagram was built from observed data. We calculated
percentages (transition frequency between behavioral acts) from the overall
number of cases.
Ant reaction to the coccinellid larvae
We conducted observations on eight B. micrantha supporting both
Libyaspis colonies on the branches and Caternaultiella
associated with C. brutus at the base of the trunk. The behavior of
the C. brutus workers when faced with the coccinellid last instar
larvae were recorded (1) at the base of the trunk, near the shelters (less
than 15 cm) where C. brutus attended Caternaultiella; (2)
when the ladybirds attacked the attended nymphs of Caternaultiella
outside the shelters; and (3) on the other parts of the trees.
To know whether attacked Caternaultiella nymphs attract their attending C. brutus workers, we conducted the following experiment at night. Nymphs belonging to clusters situated outside the shelters were excited during 10 s by rubbing the back part of their body with the tapered end of a graminaceous stalk. As a control, the end of a graminaceous stalk was rubbed against the bark of the trees in a zone situated close to clusters of bugs attended by C. brutus workers. We noted each time the number of C. brutus workers that approached and reached the excited bugs (experiment 1) or the brushed zone of the bark (control) among those situated within a radius of 20 cm (mostly occupied in attending bugs). We conducted the same experiment (experiment 2) with nymphs of Caternaultiella that we transported in petri dishes, then introduced onto the base of trees occupied by three C. brutus colonies that did not attend this plataspid; they were always accepted. In each case, 20 trials were conducted.
Statistical comparisons were made using the Kruskall-Wallis one-way test
with post hoc pairwise comparisons (Statistix 4.1 software) for comparisons of
aggressive behavior and Fisher's Exact tests (StatXact 3.1 software) for
comparisons between attractiveness to attending C. brutus workers of
clusters of bugs. We adjusted appropriate probabilities for the number of
simultaneous tests using the sequential Bonferroni procedure
(Rice, 1989).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Predatory behavior of the coccinellid last-instar larvae and reactions of the prey
A foraging ladybird larva always detected a prey by contact both under natural conditions (102 observed cases) and in the petri dishes (233 cases). Consequently, we did not note a significant difference from a random distribution during choice tests between nymphs of the two plataspid species in the petri dishes, as they attacked the Libyaspis nymphs first 11 times, and the Caternaultiella nymphs 15 times (p =.39).
Immediately after contact, the ladybirds tried to slide their long forelegs under the Libyaspis nymphs, then lifted and bit them on the ventral face (Figures 1 and 2A). Large, last-instar nymphs were never turned over, but smaller, third-instars were turned over in 42.1% of the cases (p <.001). In reaction to the attack, the Libyaspis nymphs pressed their body tightly against the plant surface, so that the hypertrophied lateral sides of the tergits came into contact with the substrate. This behavior, called "cowering" in coleopterans, permitted 27.8% of the last instars to escape the attack, but smaller nymphs were killed in all cases (p <.001; Figure 2A). We noted that, under natural conditions, cowering second- and third-instar nymphs are able to escape the attack of smaller ladybird larvae, and the relationship between prey and predator size thus plays an important role.
|
The Caternaultiella nymphs, which do not have hypertrophied extremities of the tergits, tried to escape at contact with the ladybird larvae. Only 12.9% of the last-instar and 3.3% of the third-instar nymphs succeeded in escaping (n.s.; p =.095). The behavioral sequences were more complex than for the Libyaspis, as the ladybirds, depending on the cases, slid their forelegs under the prey then lifted them or grasped and held them against their mouthparts, then bit them on the ventral face (Figures 1 and 2B).
Reactions of the plataspid nymphs of a cluster when one of their
congeners is attacked
When a ladybird attacked a Libyaspis nymph, three situations were
recorded: congeners located within a radius of about 10 cm cowered and folded
their antennae, while apparently continuing to suck sap (76.6% of 111 cases);
the cluster broke up while only certain individuals cowered (21.6%); or only
one individual left the cluster (1.8%).
In clusters of Caternaultiella composed only of nymphs, the congeners of a nymph attacked by a ladybird fled, and the cluster disintegrated (100% of the cases; n = 12). In clusters surrounded by adults, the reactions were more complex (n = 11). While the adults stayed in place but cowered, nymphs situated at the periphery indifferently stayed in place, moved toward the center of the cluster, fled, or dropped.
Reactions of C. brutus workers toward ladybirds
We observed a graded aggressiveness of C. brutus among the three
behaviors of ladybirds (foraging on tree branches, approaching shelters, or
attacking a Caternaultiella nymph;
Figure 3). The different
behaviors recorded fall into three groups: "nonaggressive
behavior" (ignore; swerve; antennate with mandibles closed),
"intimidation" (antennate with mandibles open; immobilization with
mandibles open and antennae folded backward; immobilization with mandibles
open followed by a back-and-forth movement of the body), and "aggressive
behavior" (biting; venom spraying).
|
When encountering ladybirds by chance on the tree branches, the C. brutus workers mostly ignored them or swerved (Figure 3A). Their aggressiveness was greater when a ladybird approached the shelters where workers attended Caternaultiella, with even cases of biting or venom spraying (Figure 3B). Nest mates were recruited in 57.6% of the cases (n = 33). As a result, the ladybirds turned back. Outside the shelters, Caternaultiella nymphs attacked by a ladybird triggered the disintegration of the cluster of congeners, while attracting one to three C. brutus workers (25 cases observed). Among the 78 workers situated within a radius of 20 cm of such an attack, 41 (52.6%) were attracted, while the others (47.4%) did not react. Attracted workers approached mandibles open; most of them bit the ladybirds, then cleaned their antennae and mandibles during more than one minute, or sprayed venom from a short distance (Figure 3C). All the ladybirds abandoned the prey as well as the area, but 76% of the bitten bugs died. A global statistical comparison resulted in a significant difference indicating that the C. brutus workers react differently according to the situation. The comparisons between the three groups also resulted in significant differences, confirming the graded aggressiveness of the workers according to the situation.
Reaction of C. brutus workers when a
Caternaultiella nymph was artificially excited
When a Caternaultiella individual was artificially excited with
the end of a graminaceous stalk, 41.9% of the workers within a radius of 20 cm
were attracted and arrived with mandibles open, while most of the other nymphs
flew away and the cluster disintegrated (experiment 1: 20 trials, 62 workers
within a radius of 20 cm). Rubbing a graminaceous stalk on the tree bark in a
similar manner as during the excitation of the bugs (about the same distance
from the attending workers as in experiment 1) did not trigger a reaction from
the bugs and only a slight one from the ants (12.7% of the workers attracted;
control group: 20 trials; 63 workers within a radius of 20 cm), resulting in a
significant difference with experiment 1 (Fisher's Exact tests and sequential
Bonferroni procedure: p <.001). The excitation of
Caternaultiella nymphs introduced into the hunting area of C.
brutus colonies that did not previously attend this plataspid (experiment
2: 20 trials; 59 workers within a radius of 20 cm) only triggered slight
reactions from the naive workers (16.9% of the workers). We therefore noted a
nonsignificant difference with the control group (p =.61) and a
significant difference with experiment 1 (p <.01).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Predatory behavior of the coccinellid last-instar larvae and reactions of the prey
This study presents a new illustration of the evolutionary arms race between prey and predators. The prey develops defenses including both morphological (hard dorsal cuticle and hypertrophied tergit extremities of Libyaspis) and behavioral traits (cowering), while the specialized predator foils these defenses thanks to a morphological adaptation (hypertrophied forelegs) coupled with a behavior adjusted to the situation (forelegs that the ladybird slides under the Libyaspis nymphs in order to lift them, then biting their ventral face). The proximity of the ventral nervous chain permits the injected saliva to act quickly, as we never observed nymphs struggling when bitten. Note that cowering is frequently cited as a defensive behavior in adult coleopterans that are protected by their thickened elytra, on which enemies' mandibles slip (Jiggins et al., 1993
;
Völkl, 1995).
The parsimony of the behavioral sequences when capturing Libyaspis
versus Caternaultiella nymphs
(Figure 2) can be interpreted
as the result of evolutionary processes for a specialized predator well
adapted to capture its essential prey. Sliding the forelegs under the prey
body is unnecessary for the capture of alternative prey, but only certain
last-instar A. tetrasticta larvae grasped these alternative prey,
while others behaved as when confronted with essential prey. The long forelegs
are useful in this behavior, and the sequence, grasping/salivary injection, is
strongly reminiscent of the behavior of several water bugs
(Cloarec, 1974). Moreover, even
when the lady-birds did slide their forelegs under the prey, in most cases
they did not then lift the prey, as this behavior was unnecessary. As a
result, the behavioral flexibility of the ladybirds concerned only a part of
the tested individuals; the others had a tendency to react in a stereotypical
manner.
Reactions of the plataspid nymphs of a cluster when one of their
congeners is attacked
Although living in a group increases the risk of being detected, it is
generally considered a selective advantage against generalist predators
because it dilutes predation pressure and increases vigilance and defense
(McEvoy, 1979). In our study,
an attacked nymph emitted an alarm pheromone (see
Dolling, 1991), triggering
antipredator reactions from the congeners situated in the proximity in both
plataspid species, but these reactions differed. The Libyaspis nymphs
cowered in the large majority of the cases. As a result, they remained
motionless even when one of their congeners was attacked and killed. This
permitted the population cohesion to be maintained and so provided a better
defense against further attacks. For example, cowering was sufficient against
ant aggressiveness when we installed Libyaspis colonies on
Bridelia for the first time in Yaoundé, and shortly thereafter
the ants definitively ignored them. In contrast, the reaction of the
Caternaultiella nymphs consisted of fleeing when a congener was
attacked by a ladybird, making the clusters disintegrate. In heteropteran
nymphs, the first of the three pairs of abdominal glands secretes a chemical
attractive to conspecifics and causes aggregations to form, while the second
and third secrete a chemical that causes aggregations to disperse
(Dolling, 1991). This suggests
that the two plataspid species used different glands when under attack by
ladybirds, leading to different reactions from their congeners.
Response of attending ants in the case of
Caternaultiella
Based on available evidence, we believe that the Caternaultiella
alarm pheromone, acting as an allomone in this case, is attractive to ants, as
has been recorded in aphids (Nault et al.,
1976). Moreover, the action of this product is the result of a
kind of learning on the part of the ants rather than due to similarities in
the chemical structure between the alarm pheromones of the bug and the ant, as
naive workers did not respond to stimulated nymphs. Learning in ants is well
documented and concerns different behaviors
(Corbara and Dejean, 2000;
Hölldobler and Wilson,
1990; Johnson et al.,
1994; Orivel et al.,
1997).
Reactions of C. brutus workers toward foraging coccinellid
larvae
As known for ant-attended aphids and scales
(Bradley, 1973;
Cudjoe et al., 1993;
Jiggins et al., 1993;
Nault et al., 1976;
Stadler, 1991;
Sudd, 1987), we noted a graded
aggressiveness in C. brutus toward A. tetrasticta larvae
according to the situation. When encountering A. tetrasticta on tree
branches, C. brutus workers ignored them or swerved. As almost all of
the perennial plants of equatorial countries are occupied by ants (see
Dejean et al., 2000c), we
hypothesize that the A. tetrasticta larvae are rather repellent to
ants, permitting them to share trees with ants, as reported for other ladybird
species (Sloggett and Majerus,
2000; Sloggett et al.,
1998). Moreover, A. tetrasticta probably benefits from
the ants' presence, as the latter ignored their pupae (see also
Attygale et al., 1993;
Völkl, 1995), while they
attacked the pentatomid predator of these pupae (Orivel et al., personal
observations). This situation differs from that of ladybirds that use
camouflage to enter shelters where ants attend aphids
(Völkl, 1995), as C.
brutus workers are aggressive enough to force the ladybirds that approach
the shelters where they attend Caternaultiella to flee. In this case,
C. brutus workers can even bite the ladybirds or spray venom on them,
but these behaviors are more frequent when the ladybirds attack
Caternaultiella nymphs developing outside the shelters. In these
cases, the release of a fluid from the large pores of the abdominal segments
of coccinellid larvae deters the ants (see also
Dejean, 1988;
de Jong et al., 1991;
Jiggins et al., 1993).
In conclusion, we have shown here that the ladybird A. tetrasticta
is specialized in plataspid heteropteran predation. Although morphologically
and behaviorally adapted to catch prey that cower, such as Libyaspis
nymphs, its behavioral flexibility allows for the capture of alternative prey
that escape, such as Caternaultiella nymphs. Moreover, this study
permitted us to confirm that the relationship between C. brutus and
Caternaultiella is truly mutualistic, as the attended plataspid is
protected in this case against a predator (ant protection against a parasitoid
wasp was demonstrated by Gibernau and
Dejean, 2000). This protection seems efficacious, so that the
predatory pressure of A. tetrasticta on ant-attended
Caternaultiella is limited to the period of proliferation of this
bug.
|
ACKNOWLEDGEMENTS |
|---|
We are especially indebted to William R. Dolling (Brook Farm, Elstronwick, Hull, UK) for the identification of the Plataspidae and for providing helpful information on their biology. We are grateful to Amougou Akoa (University Yaoundé I, Cameroon), Barry Bolton, and Robert D. Pope (The Natural History Museum, London), for the identification of the plants, the Formicidae, and the Coccinellidae, respectively. This work was supported by the French Ministry of Co-operation (Project CAMPUS 108/CD/90).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Attygale AB, McCormick KD, Blankespoor CL, Eisner T, Meinval J, 1993. Azamacrolides: a family of alkaloids from the pupal defense secretion of a ladybird beetle (Epilachna varivestis). Proc Natl Acad Sci USA 90: 5204-5208.
Bradley GA, 1973. Effect of Formica obscuripes (Hymenoptera: Formicidae) on the predator-prey relationship between Hyperaspis congressis (Coleoptera: Coccinellidae) and Toumeyella numismaticum (Homoptera: Coccidae). Can Entomol 105: 1113-1118.
Cloarec A, 1974. A study of the postural variations in the forelegs of Ranatra linearis (Insect Heteroptera). Behaviour 48: 89-110.
Corbara B, Dejean A, 2000. Adaptive behavioral flexibility of the ant Pachycondyla analis (= Megaponera foetens) (Formicidae: Ponerinae) during prey capture. Sociobiology 36: 465-486.
Corbara B, Orivel J, Cerdan P, Dejean A, 2001. Thalassa saginata (Coccinellidae: Hyperastinae), a myrmecophylous ladybird associated with the arboreal ant Dolichoderus bidens (Formicidae: Dolichoderinae). Naturwissenschaften, in press.
Cudjoe AR, Neuenschwander P, Copland MJW, 1993. Interference by ants in biological control of the cassava mealybug Phenacoccus manihoti (Hemiptera: Pseudococcidae) in Ghana. Bull Entomol Res 83: 15-22.
Dejean A, 1988. Memory effect on predatory behaviour of Odontomachus troglodytes (Formicidae, Ponerinae). Behaviour 107: 132-137.
Dejean A, Bourgoin T, Orivel J, 2000a. Ant defense of Euphonyarthex phyllostoma (Homoptera: Tettigometridae) during trophobiotic associations. Biotropica 32: 122-119.
Dejean A, Gibernau M, 2000. A rainforest ant mosaic: the edge effect. Sociobiology 35: 385-401.
Dejean A, Gibernau M, Bourgoin T, 2000b. A new case of trophobiosis between ants and Heteroptera. Comp R Acad Sci Paris 323: 447-454.
Dejean A, McKey D, Gibernau M, Belin M, 2000c. The arboreal ant mosaic in a Cameroonian rainforest. Sociobiology 35: 403-423.
de Jong PW, Holloway GJ, Brakefield PM, de Vos H, 1991. Chemical defense in ladybird beetles (Coccinellidae). II. Amount of reflex fluid, the alkaloid adaline and individual variation in defense in 2-spot ladybirds (Adalia bipunctata). Chemoecology 2: 15-19.
Dixon AFG, 2000. Ladybirds as predators. Cambridge: Cambridge University Press.
Dolling WR, 1991. The Hemiptera. Oxford: Natural History Museum Publications, Oxford University Press.
Evans EW, Stevenson AT, Richards DR, 1999. Essential versus alternative foods in insect predators: benefits of a mixed diet. Oecologia 121: 107-112.
Gibernau M, Dejean A, 2000. Ant protection of a Heteroptera trophobiont against a parasitoid wasp. Oecologia 126: 53-57.
Hargreaves H, 1925. Annual report of the Department of Agriculture, Entebe, Uganda.
Hodek I, 1973. Biology of Coccinellidae. The Hague: W. Junk.
Hölldobler B, Wilson EO, 1990. The ants. Cambridge: Harvard University Press.
Jiggins C, Majerus MEN, Gough U, 1993. Ant defense of colonies of Aphis fabae Scopoli (Hemiptera: Aphididae), against predation by ladybirds. Br J Entomol Nat Hist 6: 129-137.
Johnson RA, Rissing SW, Killeen PR, 1994. Differential learning and memory by co-occurring ant species. Insectes Soc 41: 165-177.
Kenne M, Schatz B, Dejean A, 2000. Hunting strategy of a generalist ant species proposed as a biological control agent against termites. Entomol Exp Appl 94: 31-40.
Majerus MEN, 1989. Coccinella magnifica (Redtenbacher): a myrmecophilous ladybird. Br J Entomol Nat Hist 2: 97-106.
Malhotra CP, Krishnaswami S, 1962. Coptosoma ostensum Dit.—a pentatomid pest of Palas (Butea monosperma) with notes on its coccinellid predator Synia melanaria Var. Rougeti Muls. and egg parasite Telenomus species. Ind Forester 88: 231-237.
Maschwitz U, Fiala B, Dolling WR, 1987. New trophobiotic symbioses of ants with South East Asian bugs. J Nat Hist 21: 1097-1107.
McEvoy PB, 1979. Advantages and disadvantages to group living in treehoppers (Homoptera: Membracidae). Misc Publ Entomol Soc Am 84: 1-13.
Nault LR, Montgomery ME, Bowers WS, 1976. Ant-aphid
associations: role of aphid alarm pheromone. Science
192: 1349-1351.
Orivel J, Errard C, Dejean A, 1997. Ant gardens: interspecific recognition in parabiotic species. Behav Ecol Sociobiol 40: 87-93.
Pope RD, 1988. A revision of the Australian Coccinellidae (Coleoptera). Part 1: Subfamily Coccinellidae. Invert Taxon 2: 633-735.
Rice WR, 1989. Analyzing tables of statistical tests. Evolution 43: 223-225.[ISI]
Sloggett JJ, Majerus MEN, 2000. Aphid-mediated coexistence of ladybirds (Coleoptera: Coccinellidae) and the wood ant Formica rufa: seasonal effects, interspecific variability and the evolution of a coccinellid myrmecophile. Oikos 89: 345-359.
Sloggett JJ, Wood RA, Majerus MEN, 1998. Adaptation of Coccinella magnifica Redtenbacher, a myrmecophilous coccinellid, to aggression by wood ants (Formica rufa group). I. Adult behavioral adaptation, its ecological context and evolution. J Insect Behav 11: 889-904.
Stadler B, 1991. Predation success of Coccinella septempunctata when attacking different Uroleucon species. In: Behaviour and impact of Aphidophaga (Polgar L, Chambers R, Dixon AFG, Hodek I, eds). The Hague: SPB Academic; 265-271.
Sudd JH, 1987. Ant aphid mutualism. In: Aphids—their biology, natural enemies and control (Minks AK, Harrewijn P, eds). Amsterdam: Elsevier; 355-365.
Völkl W, 1995. Behavioral and morphological adaptations of the coccinellid, Platynaspis luteorubra for exploiting ant-attended resources (Coleoptera: Coccinellidae). J Insect Behav 8: 653-670.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||