Content uploaded by Jean-Louis Hemptinne
Author content
All content in this area was uploaded by Jean-Louis Hemptinne
Content may be subject to copyright.
REVIEW Special feature: predatory ladybirds: individuals,
populations, and interactions
Interactions between ants and aphidophagous
and coccidophagous ladybirds
Michael E. N. Majerus ÆJohn J. Sloggett Æ
Jean-Franc¸ois Godeau ÆJean-Louis Hemptinne
Received: 13 December 2005 / Accepted: 17 August 2006 / Published online: 16 November 2006
ÓThe Society of Population Ecology and Springer 2006
Abstract Aphidophagous and coccidophagous cocc-
inellids come into conflict with homopteran-tending
ants for access to food. Antagonistic interactions be-
tween coccinellids and ants may be competitive or non-
competitive. Competitive interactions occur when
coccinellids attack aphids or coccids that are being
tended by ants for honeydew. Non-competitive inter-
actions include all interactions away from ant-tended
homopteran colonies. We here review observations
and studies of such interactions. We note that most
competitive interactions occur at times when untended
aphids/coccids are scarce. We describe the chemical
and physical defences that coccinellids use against ant
aggression and consider whether these have evolved as
general anti-predator deterrents or specifically in re-
sponse to ants. Myrmecophilous coccinellids are then
considered, with particular focus on the two most
studied species, Coccinella magnifica and Platynaspis
luteorubra. We note that the myrmecophily of the two
species has the same adaptive rationale—to enable the
ladybirds to prey on ant-tended aphids at times of
aphid scarcity—but that it is based on different traits to
facilitate life with ants. Finally, we consider the role of
ants in the evolution of habitat specialisation in some
coccinellids.
Keywords Coccinellidae Myrmecophily Aphids
Coccids Optimal forage theory Chemical defence
Introduction
Insects that are associated with ants are called myr-
mecophilous (myrmex = ant: Greek). Myrmecophilous
aphids and coccids show behavioural and structural
modifications to life with ants. When an ant encounters
such an insect, it usually strokes it with its antennae.
This induces the aphid or coccid to suppress its usual
defensive behaviour of kicking out, running away,
dropping off the plant or clamping down. Instead, it
raises its abdomen and exudes droplets of honeydew,
which the ants then imbibe. Ants gain food from the
association, for honeydew is rich in carbohydrates and
also contains amino acids, amides, proteins, minerals
and B-vitamins (Way 1963; Carroll and Janzen 1973;
Ho
¨lldobler and Wilson 1990). At times, ants also gain
protein, by preying on aphids or coccids. Benefits to
the aphids or coccids include improved hygiene via
removal of caste skins, dead aphids and honeydew
(Way 1954; Banks 1958; Seibert 1992); direct increases
in development rate, adult body size, fecundity and
M. E. N. Majerus (&)
Department of Genetics, University of Cambridge,
Downing Street, Cambridge CB2 3EH, UK
e-mail: m.majerus@gen.cam.ac.uk
J. J. Sloggett
Department of Entomology, University of Kentucky,
S-225 Agricultural Science Center North, Lexington,
KY 40546-0091, USA
J.-F. Godeau
Biodiversity Research Centre,
Universite
´Catholique de Louvain,
Place Croix du Sud, 4-5,
1348 Louvain-la-Neuve, Belgium
J.-L. Hemptinne
Evolution et Diversite
´Biologique,
Ecole Nationale de Formation Agronomique,
UMR 5174 CNRS/Universite
´Toulouse III/ENFA,
2, route de Narbonne, BP 22687,
31326 Castanet-Tolosan Cedex, France
123
Popul Ecol (2007) 49:15–27
DOI 10.1007/s10144-006-0021-5
reproductive rate (El-Ziady and Kennedy 1956; Banks
1958; El-Ziady 1960); and protection from enemies
(Bartlett 1961; Banks 1962; Jiggins et al. 1993).
By protecting homopterans from predators and
parasitoids, ants come into conflict with such species
(Rosen 1990; Dixon 1998). Evolutionary and ecolog-
ical responses of parasitoids to antagonism from ants
include more rapid ovipositing (Bartlett 1961),
avoidance behaviour (Vo
¨lkl 1997) and various
chemical adaptations (Liepert and Dettner 1996;
Vo
¨lkl 1997). Evolutionary and ecological responses of
homopteran predators to ant attendance of their prey
has received less attention. Although many attributes
of aphidophages and coccidophages have been sug-
gested to result from interactions with Homoptera-
tending ants, evidence supporting such suggestions is
sparse, scattered back over a century, and is often
contradictory.
In this report, we briefly review literature on inter-
actions between one group of homopteran preda-
tors—the ladybirds (Coccinellidae)—and ants. We
consider when and why ladybirds feed on ant-tended
Homoptera, describe two case studies of myrmecoph-
ilous coccinellids and speculate on the evolution of
responses to ants and the evolution of myrmecophily.
Interactions between ants and ladybirds
Ant–ladybird interactions are of three types. First, and
most importantly, ants that tend Homoptera compete
with aphidophagous or coccidophagous ladybirds for
resources. Second, ladybirds may feed on ants, al-
though only one ladybird is known to specialise on ants
(Harris 1921). Third, ants may prey on ladybirds. The
most useful separation of these interactions is into
competitive and non-competitive.
Competitive ant–ladybird interactions involving
homopterans
There is considerable evidence that Homoptera-tend-
ing ants are more aggressive towards coccinellids in the
vicinity of tended colonies than elsewhere. This has
been described as ownership behaviour (Way 1963).
This aggression is aimed at both adult and larval
coccinellids. Adults are usually chased from homop-
teran colonies (Bradley 1973; McLain 1980; Itioka and
Inoue 1996; Sloggett 1998), while soft-bodied larvae
may be picked up and carried away from the colony,
dropped off the plant, or killed (Bradley 1973; Vinson
and Scarborough 1989; Bach 1991; Jiggins et al. 1993;
Sloggett and Majerus 2003).
Studies comparing the density of ladybirds in the
presence and absence of ants have usually shown that
ants reduce ladybird numbers on ant-tended colonies
of both aphids and coccids (natural presence/absence
of ants: Mariau and Julia 1977; McLain 1980;Vo
¨lkl and
Vohland 1996; ant absence due to artificial barriers:
Bradley 1973; Reimer et al. 1993; Itioka and Inoue
1996; or poisoning: Mariau and Julia 1977; Jutsum et al.
1981). Exclusion of coccinellids from homopteran col-
onies by ants is beneficial to both tended aphids (Banks
1962; Mariau and Julia 1977; Reimer et al. 1993) and
tended coccids (Bradley 1973).
Coccinellid predation of ant-tended Homoptera
A few coccinellids are considered myrmecophilous.
While such species habitually live near ant nests, most
non-myrmecophilous coccinellids only feed on ant-
tended Homoptera when untended Homoptera are
scarce (Sloggett and Majerus 2000a). That said, dif-
ferences in the size, aggressiveness and density of
tending ants, and in the size, behaviour and defensive
capabilities of coccinellids, undoubtedly affect the level
and outcome of ladybird/ant interactions.
The most detailed work on coccinellid predation of
tended and untended Homoptera has involved aphi-
dophagous species in temperate regions. Aphidopha-
gous coccinellids usually breed in periods of aphid
abundance, when adults are feeding on untended prey
(Majerus 1994; Hodek 1996; Sloggett 1998). Conse-
quently, immature coccinellids (excepting myrmeco-
philes) rarely come into conflict with ants tending
aphids. Only in years with a general scarcity of aphids
will coccinellid larvae try to attack ant-tended aphids
and so be under selection for adaptations to enable
them to feed on such prey (Sloggett 1998).
In contrast, adult coccinellids come into conflict with
aphid-tending ants annually in late summer when feed-
ing up prior to overwintering. Due to aphid scarcity at
this time, they feed on alternative foods, such as pollen,
nectar, sap, honeydew, non-homopteran invertebrates
and conspecifics (Hodek 1996; Sloggett and Majerus
2000b). Some also attack ant-tended aphids (Sloggett
1998; Sloggett and Majerus 2000a). The tolerance of
adult ladybirds to ant aggression then becomes critical,
as those with little tolerance are then forced to feed on
non-homopteran food. Unfortunately, few studies allow
an assessment of the relative tolerances of different
coccinellids to ants. Most studies of the effects of ants on
coccinellids have involved a single target species of
ladybird (Bradley 1973; Itioke and Inoue 1996) or have
clumped results of all coccinellids together (Banks and
Macaulay 1967; Bristow 1984). However, two studies
16 Popul Ecol (2007) 49:15–27
123
showed that coccinellids vary in their tolerance to ants,
even if the myrmecophiles are excluded.
DeBach et al. (1951) observed that 66% of Rhizo-
bius lophanthae was found in the presence of the ant
Iridomyrmex humilis tending the coccid Aonidiella
aurantii, on citrus, but only 15% of a Chilocorus spe-
cies occurred in the same situation.
In a study in an English pine forest, the numbers of
six species of coccinellid, two types of aphid, and ant
presence or absence were monitored from spring to
autumn (Sloggett 1998; Sloggett and Majerus 2000a).
The ladybirds comprised four conifer special-
ists—Myrrha 18-guttata,Anatis ocellata,Myzia oblon-
goguttata and Harmonia 4-punctata—the generalist
Coccinella 7-punctata and the myrmecophile Cocci-
nella magnifica. The aphids were Schizolachus pineti,
which are not tended by ants, and two Cinara species,
present at two adjacent sections, one containing several
F. rufa nests (ant plot), while the other was free of
F. rufa (control plot). Cinara aphids on Pinus sylvestris
in the ant plot were regularly tended by F. rufa, while
those in the control plot were not.
Analysis of the patterns of abundance of the various
ladybirds in the two plots through the summer allows
determination of the ant-tolerance. Myrrha 18-guttata
and A. ocellata had little tolerance of ants, these spe-
cies only occurring in the ant plot after ants had dis-
appeared in September. A third species, H. 4-punctata,
exhibited low ant tolerance, being much less abundant
in the ant plot than in the control plot. Coccinella
7-punctata showed a similar pattern but with a slightly
higher tolerance. M. oblongoguttata were found sig-
nificantly more in the control than in the ant plot
during early summer, when aphids were abundant.
However, once aphids became scarce, their abundance
was similar in the two. That it increased in the ant plot
once aphids became scarce strongly suggests that it
moved into this area to feed on Cinara aphids even
though these were ant-tended. M. oblongoguttata are
highly specialised in their diet, only breeding when
feeding on conifer aphids, particularly Cinara spp.
(Majerus 1993). This specialisation towards a few aphid
species, most of which elicit ant attendance, will have
imposed selection pressure for M. oblongoguttata to be
ant tolerant. The main defence observed was physical,
the ladybird dropping the elytra down to the substrate
on the side being assailed by ants. Occasionally,
M. oblongoguttata have been observed to run from
ants, but rarely dropped from the pines and was not
observed to fly away. The sixth species of ladybird,
C. magnifica, was more abundant in the ant area than
in the area lacking ants throughout the study, con-
firming its myrmecophile status (see below).
Larvae of five species were found on P. sylvestris,
those of C. 7-punctata not being found. No larvae of
M. 18-guttata were found in the ant area. Larvae of
A. ocellata,H. 4-punctata and M. oblongoguttata were
much more abundant in the ant-free area than in the ant
area. Larvae of C. magnifica were confined to the ant
area.
From these results, Sloggett and Majerus (2000a)
drew up an order of ant tolerance for these six ladybirds:
M. 18-guttata +A. ocellata <H. 4-punctata <C.
7-punctata <M. oblongoguttata <C. magnifica. They
concluded that C. magnifica is a true myrmecophile,
while M. oblongoguttata has some defence against ants,
both as an adult and as larva. Adult C. 7-punctata
coexists with F. rufa at moderate levels when aphids are
scarce, but does not breed in the presence of F. rufa.
Most work on ladybird–ant interactions have in-
volved ant-tended aphids. However, the levels of pre-
dation of ant-tended aphids and ant-tended coccids by
ladybirds might be quite different. As most coccinellids
mainly attack ant-tended Homoptera when non-tended
Homoptera are scarce, the probability of finding un-
tended Homoptera becomes critical. Several factors
may cause differences in the likelihood that coccinel-
lids will encounter ant-tended aphids when compared
with ant-tended coccids. First, at higher latitudes,
aphids are much more common relative to coccids than
in the tropics. Ant diversity and abundance is much
greater in the sub-tropics and tropics than in more
temperate climes. Thus, untended colonies of coccids
will be less common than untended colonies of aphids.
Second, and conversely, aphids are renowned for the
ephemerality of their colonies, particularly in seasonal
climes: coccids less so (Dixon 2000). In part, aphid
ephemerality is due to eradication of colonies by pre-
dators and parasitoids, although other factors, such as
the concentration of soluble sap available in plant sap
(Dixon 1970; Strong et al. 1984), are also important.
The ‘‘boom and bust’’ cycle seen in aphids means that
their predators often face a dearth of non-ant-tended
aphids that have been protected from this predation/
parasitoid pressure. Ant-tended colonies of aphids
persist for longer than untended colonies of the same
species (Addicott 1979; Bristow 1984; Mahdi and
Whittaker 1993; Sloggett and Majerus 2000a). Foraging
theory predicts that aphidophagous coccinellids are
most likely to feed on ant-tended aphids when un-
tended aphids are scarce, i.e., when untended aphids
are of greater relative value (Stephens and Krebs
1986). In temperate regions, untended aphids become
scarce in late summer, while ant-tended colonies often
remain abundant at this time (Mahdi and Whittaker
1993; Sloggett and Majerus 2000a).
Popul Ecol (2007) 49:15–27 17
123
It is not clear whether overall aphidophagous cocc-
inellids or coccidophagous coccinellids are under
greater pressure to feed on ant-tended prey. What is
clear is that the pressure will vary with both coccinellid
species and prey species. Moreover, this pressure will
vary greatly in seasonal habitats (sensu Southwood
1977), which will be more pronounced in temperate
regions than in the tropics. Work on the relative
availabilities of untended and tended aphids and
coccids through the year in a variety of climate zones is
urgently needed, and may shed light on both similari-
ties and differences in the interactions between coc-
cidophagous and aphidophagous coccinellids and
Homoptera-tending ants.
Non-competitive interactions
Non-competitive interactions include all those away
from ant-attended homopteran colonies, plus instances
of predation of ladybirds by ants (or the reverse). Such
interactions are important as they influence habitat
preferences and ladybird distributions within an envi-
ronment. Away from homopteran colonies, ants that
encounter ladybirds either attack them or ignore them,
depending usually on the species of ant. Thus, several
ants that attack coccinellids in the vicinity of tended
homopterans, including Lasius niger (El-Ziady and
Kennedy 1956; Banks 1962), Formica fusca (Rathcke
et al. 1967), I. humilis (Dechene 1970) and Myrmica
ruginodis (Jiggins et al. 1993), are indifferent to lady-
birds elsewhere. Conversely, some ants that are prey
on insects attack ladybirds whenever they encounter
them, and so exclude many coccinellids from their
forage range. Examples are few because most empiri-
cal evidence is based on introducing coccinellids to
captive, starved ant colonies, and interactions do not
reflect what happens in the field, and because it is
difficult to disentangle the effects of ant predation of
coccinellids from ant attendance of Homoptera. Thus,
Hays and Hays (1958) found that captive, starved
Solenopsis invicta would kill and eat five species of
coccinellid, yet Wilson and Oliver (1969) found only
one coccinellid among 4,056 prey items taken by this
ant, and Sterling et al. (1979) found that S. invicta
presence did not reduce coccinellid numbers in cotton
fields. Such difficulties mean that separating ant-cocc-
inellid interactions into competitive and non-competi-
tive may be overly simplistic. However, despite this
artificiality, it is likely that highly aggressive predatory
ants have a greater effect on coccinellid distributions in
a habitat than do ants that only attack ladybirds near
homopteran colonies.
Coccinellid defences against ants
The tolerance of ladybirds to ants depends, at least in
part, on the defensive capabilities of the ladybirds.
Coccinellids use various mechanisms when faced with
ant aggression (Pasteels et al. 1973; Richards 1980,
1985; Majerus 1994). These defences may be behavio-
ural, physical or chemical, with some being shown at
specific periods of the life cycle, and others shown at
both adult and immature stages.
Behavioural defences
Most coccinellids exhibit defensive behaviours when
attacked by ants. Most commonly, larvae escape by
running away or dropping to the ground, while adults
may fly (Banks 1962; Bradley 1973; Itioka and Inoue
1996) in addition to using these tactics. For adults, an
alternative to fleeing is to ‘‘clamp down’’, retracting
their legs under the body, pulling their heads close to
the thorax and attaching firmly to the substrate
(Bradley 1973; Jiggins et al. 1993; Majerus 1994).
Members of the sub-family Chilocorinae have a very
flat ventral surface and a lip around the elytral edge, so
that the contact made when they clamp is very tight
and prevents ants from gaining access to the vulnerable
ventral surface of the ladybird. Conversely, many
Coccinellinae species do not clamp down completely,
but adopt a rolling motion, dropping the side being
attacked to make close contact with the substrate
(Jiggins et al. 1993; Sloggett 1998).
Many coccinellid pre-pupae and pupae can rapidly
raise their anterior end in response to tactile stimuli.
This ‘‘pupal flicking’’ behaviour (Majerus 1994) may be
repeated many times. Eisner and Eisner (1992) sug-
gested that this is a defence against ants, with the joints
between abdominal segments acting as ‘‘gin-traps’’ that
damage ant appendages. However, a more likely
explanation of this behaviour is that it reduces ovipo-
sition by pupal parasitoids, such as scuttle flies (Dip-
tera: Phoridae) (Disney et al. 1994).
It is unclear whether these behavioural defences
evolved as specific responses to ant aggression or are
general anti-predator/parasitoid devices, although
Sloggett (1998) suggests that some of these behaviours
are more extensively developed in species that
encounter ants frequently.
Physical defences
The chorion of ladybird eggs is relatively thin and is
unlikely to be effective against ant mandibles. The
exoskeleton of coccinellid larvae is soft and easy to
18 Popul Ecol (2007) 49:15–27
123
pierce. However, many coccinellid larvae are covered
by spines (Richards 1980), which may provide some
protection against ant attack, although this has not
been demonstrated experimentally (Sloggett 1998).
The exoskeletons of coccinellid pupae are relatively
hard and, although not impregnable to predators or
parasitoids, will provide some protection against ant
attack. Moreover, except in the Coccinellinae, Sti-
cholotinae and a few species from other sub-families,
the pupa gains some protection from the final larval
skin (Richards 1980), which is not shed back, but
simply splits along the dorsal mid-line during pupation.
This additional layer, which is sometimes spiny or
waxy, should be considered a general defensive adap-
tation.
Some coccinellid larvae are covered on the ventral
surface by a network of wax filaments. Pope (1979)
proposed that this wax covering was an adaptation
against ant attack. First, it may be difficult for ants to
bite into. Second, as some waxes are sticky, it may
cause ants to break off attacks to clean their mouth-
parts. The defensive efficiencies of wax coverings of
two species of Scymnus were examined by Vo
¨lkl and
Vohland (1996). They showed that mortality of normal
larvae (waxy) of S. nigrinus and S. interruptus, caused
by attacks from Formica polyctena and L. niger,
respectively, was lower than that of larvae from which
the wax had been removed. Although some normal
larvae were killed, the ants that attacked these larvae
frequently broke off attacks to clean their mouthparts.
Moreover, Vo
¨lkl and Vohland (1996) found that
numbers of S. nigrinus larvae were significantly higher
closer to F. polyctena than in its absence, while num-
bers of S. interruptus were similar in the presence and
absence of L. niger.
A third adaptive function of larval wax may be mi-
metic. Some species with wax coverings feed on mealy
aphids and resemble their prey closely. It is thus pos-
sible that mealybug-tending ants ignore such coccin-
ellid larvae because the larvae are not recognised as a
threat to the homopterans. Support comes from
observations of Cryptolaemus montrouzieri larvae
being ignored by Pheidole megacephala when on col-
onies of waxy mealybugs tended by this ant, but at-
tacked by the same ant on tended colonies of the
waxless Coccus viridis (Bach 1991). The wax coverings
of coccinellids are secreted by the larvae themselves.
Interestingly, some chrysopid (Neuroptera), larvae
harvest wax from their homopteran prey and stick it to
their dorsal surface (Eisner et al. 1978). These larvae
frequently feed on ant-tended aphids. As the wax
coverings of the aphids and chrysopid larvae are
indistinguishable, the ants do not attack the larvae.
Some coccinellid pupae also have wax coverings.
The larva of Scymnodes lividigaster smears wax onto
the surface of the substrate where it attaches before
pupation (Richards 1980). The pupa that is formed is
both wax-covered and spiny. Richards (1980) has pro-
posed that both the wax smear and the pupal covering
act to deter aphid-tending ants and other predators. If
so, it is not clear whether deterrence is from the
physical barrier of the pupal covering, the texture of
the wax, its chemical composition, its colour or a
combination of these (Richards 1980; Sloggett 1998).
The pupa of Rodatus major has a very dense wax
covering, which, in addition to being a physical barrier
against ant aggression, may have a mimetic function.
The species feeds mainly on eggs of the coccid Mo-
nophlebulus pilosior, which are often tended by Irido-
myrmex ants. Richards (1985) has proposed that the
wax gives R. major larvae a resemblance to the ovisac
of this coccid.
The main physical defence of adult coccinellids
against ant attack is its hard dorsal surface. Coupled
with the clamping and rolling behaviours, the dorsal
surface provides a stout barrier to injury from ants. The
fine hairs that cover the elytra of some coccinellids may
provide additional protection against ant attack, but
this has yet to be demonstrated.
Chemical defences
Coccinellids are well known for their bright coloured
patterns, which are generally considered to be apose-
matic, advertising unpalatability (Brakefield 1985;
Majerus 1994). This unpalatability is largely chemical
in nature. Coccinellids reflex bleed, secreting a foul-
smelling, distasteful fluid from the tibio-femoral joints
of adults or the dorsal surface of larvae and pupae.
At the centre of coccinellid defence lies an array of
alkaloids (Daloze et al. 1995) and pyrazines (Moore
et al. 1990). The variety of defensive chemicals in the
coccinellids and the variation in concentrations of the
substances present in these cocktails indicate that these
insects were some of the first to use combinatorial
chemistry in their defence (Schro
¨der et al. 1998).
Many, but not all, of the defensive chemicals found
in ladybirds are synthesised by the ladybirds them-
selves (Tursch et al. 1976; Jones and Blum 1983).
However, some coccinellids also have the ability to
store and use defensive chemicals from their prey.
Coccinella undecimpunctata and Hippodamia variegata
both sequester cardiac glycosides from Aphis nerii
(Rothschild and Reichstein 1976). Hyperaspis trifur-
cata gain a major weapon by storing anthraquinone
carminic acid from its main prey, cochineal insects of
Popul Ecol (2007) 49:15–27 19
123
the genus Dactylopius. Similarly, C. septempunctata
sequester pyrolizidine alkaloids when feeding on Aphis
jacobaeae (Witte et al. 1990). In these latter two cases,
the defensive chemicals are manufactured by the plants
on which the aphids feed, so the ladybirds get these
defensive elements third hand.
The defensive chemicals of many coccinellids are
distasteful or toxic to many predators (Morgan 1896;
Pasteels et al. 1973; Brakefield 1985; Marples et al.
1989), but not to all (Muggleton 1978; Majerus 1994;
Majerus and Majerus 1997). The general consensus is
that reflex blood is distasteful to ants (Sta
¨ger 1929;
Happ and Eisner 1961; Pasteels et al. 1973; Sloggett
1998). Furthermore, ants contaminated by reflex blood
may have their mobility impaired by it as it dries
(Sta
¨ger 1929; Bhatkar 1982).
The extent to which adult coccinellids reflex bleed in
response to ant attacks varies between species and
circumstances. Some observers have reported that
ladybirds rarely reflex bleed, even under sustained at-
tack by ants (Marples 1993; Jiggins et al. 1993), while
others have observed ladybirds reflex bleeding readily
when attacked (Banks 1962; Bhatkar 1982). Majerus
(1994) argued that reflex bleeding is used by adult
coccinellids against ants as a last defence, when other
defences, including fleeing, have failed. Reflex bleed-
ing is costly due to energy expended in chemical syn-
thesis and in fluid loss (de Jong et al. 1991; Holloway
et al. 1991,1993) and is therefore only deployed when
other strategies have failed and the ladybird is in se-
vere jeopardy (Majerus 1994).
Sloggett (1998) used the cost-benefit argument to
conclude that reflex bleeding did not evolve initially as
a defence against ants. He noted that coccinellids most
often come into conflict with homopteran-tending ants
at times of ant-untended homopteran scarcity. At such
times ladybirds will have low resource reserves and so
reflex bleeding would incur a relatively high cost. It is
notable that the phytophagous coccinellid E. varivestis
reflex bleeds readily when attacked by ants (Happ and
Eisner 1961). This species synthesises a vast array of
defensive alkaloids (Eisner et al. 1986; Attygalle et al.
1993a,b; Proksch et al. 1993; Shi et al. 1997; Radford
et al. 1997). Sloggett argued that this complexity may
be a consequence of the plant diet of E. varivestis,so
that, unlike homopteran predators, it will rarely be
food limited and thus be able to devote more resources
to chemical defence and reflex bleeding.
The evidence that coccinellid larvae reflex bleed
when attacked by ants is clear (El-Ziady and Kennedy
1956; Happ and Eisner 1961; Bradley 1973; Sloggett
1998). This is because larvae are at greater risk of
suffering injury from ants than are adult ladybirds
(Majerus 1994). It may also be that as larvae usually
occur at times of prey abundance, they are less re-
source limited than adults (Sloggett 1998).
Ant aggression probably played little role in the
initial evolution of reflex bleeding in coccinellids.
However, it may have a role in shaping the precise
balance of defensive capabilities of coccinellids to a
variety of predators, parasites and pathogens. In
coccinellids that frequently encounter ants, more re-
sources may be devoted to defences against ants (and
less against other enemies) than in species that rarely
interact with ants. In addition, ants may reduce the
density of potential coccinellid predators and parasi-
toids occurring within ant forage ranges, producing
enemy-free space (Jeffries and Lawton 1984). If so,
ladybirds that commonly co-occur with ants, including
myrmecophiles, may invest fewer resources in defences
against predators/parasites that are excluded by ants
than would ladybirds that rarely occur with ants
(Sloggett 1998).
Coccinellid eggs and some pupae also have chemical
defences. The eggs of aphidophagous coccinellids
contain defensive chemicals that deter some predators
(Agarwala and Dixon 1992; Majerus 1994; Hemptinne
et al. 2005), including ants (Godeau 1997; Sloggett
1998). For example, F. rufa workers find the eggs of
C. septempunctata repellent, although they may still
destroy the eggs (Sloggett 1998). Some coccinellid
pupae (e.g., Chilocorini) reflex bleed and this blood
has some deterrent effect against ants. Pupae of
E. varivestis have a covering of glandular hairs, each
hair producing a droplet of alkaloid that is repellent to
ants (Attygalle et al. 1993a).
Myrmecophily
The extent of myrmecophily among coccinellids
A small number of coccinellids are regularly associated
with one or more species of ant and may be myrme-
cophilous (Berti et al. 1983; Sloggett 1998). These
associations are listed in Table 1, together with the
evidence on which their suggested myrmecophily is
based. Of the 11 taxa listed, myrmecophily should be
considered unproven in 4. The suggestion of myrme-
cophily in the tribe Monocorynini is based only on
antennal morphology, and records of associations be-
tween these ladybirds and ants are lacking (Sloggett
1998). In Scymnus fenderi and S. formicarius, the
possibility of myrmecophily eminates from a few re-
cords of adults and pupae being found with ants. The
myrmecophily of Hyperaspis acanthicola is based on
20 Popul Ecol (2007) 49:15–27
123
larvae being found in hollow Acaci spines abandoned
by ants and may also be unsound. For all these species,
further observations to determine the extent of asso-
ciations with ants are needed. Experimental work to
test whether these coccinellids benefit from any asso-
ciations found with ants would be even more valuable.
The myrmecophily of the remaining seven taxa is
more certain, but in some little is known of the precise
nature of the association with ants. In Brachiacantha
quadripunctata,B. ursina,Hyperaspis reppensis and
Ortalia pallens, myrmecophily may be limited to the
larvae. In each species, larvae have been found in ants’
nests feeding on ant-tended fulgorids or coccids, and in
O. pallens on the host ants. The predation of ants by
O. pallens probably results from a dietary shift after it
had developed a myrmecophilous habit (Sloggett
Table 1 Coccinellids that have been suggested as being myrmecophilous
Coccinellid Associated ant(s) Evidence of myrmecophily References
Subfamily: Coccidulinae
Tribe Monocorynini
Various species
Unknown Adults have compact antennal
clubs. Myrmecophily
unproven
Kova
´r(1996)
Subfamily: Scymninae
Scymnus fenderi Pogonomyrmex
subnitidus
One adult recorded from P.
subnitidus nest. Ant is
gramnivorous and does not
tend Homoptera.
Myrmecophily unproven
MacKay (1983) and Ho
¨lldobler
and Wilson (1990)
Scymnus formicarius Formica rufa Little known. Adults apparently
found with ants
Wasmann (1894)
Brachiacantha
quadripunctata
Lasius umbratus
Formica subpolita
(=F. camponoticeps)
Waxy larvae prey upon tended
coccids and adelgids within
ant nests. Closely related
species are probably also
myrmecophilous. Other ant
hosts are probable
Mann (1911), Wheeler (1911),
Gordon (1985), and
Montgomery and Goodrich
(2002)
Brachiacantha ursina Lasius spp. Probably the same behaviour as
B. quadripunctata
Smith (1886) and Montgomery
and Goodrich (2002)
Hyperaspis reppensis Tapinoma nigerrimum Larvae apparently feed on ant-
tended fulgorids in ants’ nests.
Adults are attacked by ants
Silvestri (1903)
Hyperaspis acanthicola Pseudomyrmex ferruginea Larvae found in hollow spines of
Acacia spp. abandoned by
ants. Myrmecophily unproven
Chapin (1966)
Ortalia pallens Pheidole punctulata Myrmecophilous larvae feed on
ants. Adult habits unknown
Harris (1921)
Thalassa saginata Hypoclinea bidens Pupae found with ants.
Chemical mimicry
demonstrated for larvae,
pupae and adults. Production
of chemical attractant.
Myrmecophily probable. Diet
unknown, hypothesis of ant
brood predation by larvae
and/or adults
Berti et al. (1983), Corbara et al.
(1999), and Orivel et al.
(2004)
Subfamily: Chilocorinae
Platynaspis luteorubra Lasius niger
Myrmica rugulosa
Tetramorium caespitum
Multiply recorded with a variety
of ant species. Larvae, and
pupae show myrmecophilous
morphology
Pontin (1959), Majerus (1994),
and Vo
¨lkl (1995)
Subfamily: Coccinellidae
Coccinella magnifica Formica rufa group All stages found with ants Donisthorpe (1919–1920),
Wasmann (1912), Majerus
(1989), Sloggett (1998),
Sloggett et al. (1998), and
Sloggett and Majerus (2003)
Adapted from Sloggett (1998)
Popul Ecol (2007) 49:15–27 21
123
1998). The larvae and pupae of Thalassa saginata de-
velop in the nests of Dolichoderus bidens (Berti et al.
1983). The larvae mimic cuticular lipids of the ants’
brood, although whether they feed on the brood is not
known (Orivel et al. 2004).
In Coccinella magnifica and Platynaspis luteorubra,
adaptations to a myrmecophilous existence are seen in
the larvae and in some other life-history stages. The
myrmecophily of both species has been studied in some
detail and consideration of these cases sheds light not
only on the ecology and evolution of myrmecophily,
but on interactions between ants and insects that prey
on Homoptera more generally.
The case of Coccinella magnifica
Coccinella magnifica is a known myrmecophile
(Sloggett et al. 2002 and references therein). It occurs
through much of the Palaearctic, its local distribution
resulting from its association with ants. In northwest-
ern Europe, it occurs in the foraging areas of ants of
the F. rufa group.
The obligate myrmecophily of C. magnifica was
confirmed in Britain during a general survey of cocc-
inellids. Samples of all seven spotted ladybirds of
appropriate size were collected at 26 sites in southern
England, without consideration of the presence or ab-
sence of F. rufa. A total of 5,971 ladybirds were col-
lected. Of these, only 49 were C. magnifica, all found
along with F. rufa, and the remainder was C. septem-
punctata (Majerus 1989).
Donisthorpe (1919–1920) placed C. magnifica and
C. septempunctata on F. rufa nests, the former being
only slightly attacked while the latter were ‘‘vigorously
assailed’’. Pontin (1959) and Majerus (1989) recorded
similar observations. Some workers have recorded that
adult C. magnifica reflex bleed freely when attacked by
ants (Donisthorpe 1919–1920), while others have ob-
served that its relative immunity to attack by ants was
rarely associated with reflex bleeding (Majerus 1989;
Jiggins et al. 1993).
Various hypotheses have been put forward to ex-
plain the low levels of aggression of ants towards C
magnifica. Possibly, C. magnifica secrete a pheromone
that deters ants (Majerus 1989; Sloggett 1998)by
advertising distastefulness or toxicity. Alternatively,
the ladybird may secrete chemicals that mimic the ants’
own scent or, possibly, the odour of aphids (Majerus
1989). Third, C. magnifica may exude a chemical that is
harmful to ants (Donisthorpe 1919–1920). Field and
laboratory studies of the interactions of C. magnifica
and other ladybirds that occur with F. rufa in conifer
and mixed woodland were used to test these hypoth-
eses (Sloggett et al. 1998; Sloggett and Majerus 2003).
Of particular note are experiments on the behaviour of
F. rufa towards C. magnifica and C. septempunctata
introduced onto ant foraging trails and ant-tended
aphid colonies. On trails, C. magnifica were attacked
occasionally, but very much less than C. septempunc-
tata. This finding is important, as Godeau et al. (2003)
have shown that C. magnifica follow ant trails to locate
aphid colonies. Moreover, on ant-tended aphid colo-
nies, C. magnifica stayed on the colony longer and were
more successful in feeding on aphids than C. septem-
punctata. Although ants attacked both species, the
degree of aggression towards C. septempunctata was
greater than that towards C. magnifica. In response to
attacks, C. septempunctata dropped off plants or flew
away significantly more often than C. magnifica.
C. septempunctata adults occasionally responded to ant
attacks by reflex bleeding, in contrast to C. magnifica
which never did. Larvae of both species were seen
to reflex bleed when attacked by ants. Although
C. magnifica larvae reflex bled much less often than
C. septempunctata, ants were more deterred from
attacking C. magnifica larvae than C. septempunctata
larvae when they reflex bled (Sloggett and Majerus
2003). The soft-bodied C. magnifica larvae were also
frequently found in situations that minimised ant
aggression, feeding on aphids dislodged onto ant trails
or on untended aphids. Indeed, C. magnifica appear to
lay eggs away from ant-tended aphids, but close to
untended aphids (Sloggett and Majerus 2003). Finally,
none of the defensive behaviours of C. magnifica in
interactions with F. rufa was unique to C. magnifica.
All were also seen in C. septempunctata, with differ-
ences between species being of the degree to which the
various behaviours were used. Thus, C. magnifica’s
defence against ants may have evolved by gradual
adaptation of C. septempunctata behaviours.
Sloggett (1998) also considered the chemistries of
C. magnifica and C. septempunctata. He showed that
dead C. septempunctata were more frequently attacked
on ant trails than were C. magnifica, whether whole
corpses, corpses without elytra or wings, or just elytra
were used. He deduced that the low level of aggression
shown by ants to C. magnifica has a chemical basis.
Analysis of the cuticular lipids of the two species
showed little difference, and little similarity to the
surface lipids of F. rufa (G. Lognay, J.J. Sloggett and
J-L. Hemptinne in Sloggett 1998). Due to the lack of
similarity between the surface lipids of C. magnifica
and F. rufa, Sloggett (1998) argued that C. magnifica’s
immunity to ant attacks was not due to chemical
mimicry of the ants. Moreover, the similarity in the
cuticular lipids of C. magnifica and C. septempunctata
22 Popul Ecol (2007) 49:15–27
123
make it improbable that C. magnifica gains immunity
by mimicking another element in the habitat. Transfer
experiments showed that C. magnifica’s defence is not
F. rufa nest specific (Sloggett 1998). Sloggett concluded
that C. magnifica’s defence is probably based on
repellent chemistry and that the chemicals involved are
alkaloids, and possibly pyrazines. Interestingly, while
the predominant alkaloids produced by most Cocci-
nella species are coccinelline and precoccinelline, those
of C. magnifica are hippodamine and convergine
(Dixon 2000; Sloggett 2005). Convergine is more
repellent to ants than coccinelline (Pasteels et al. 1973).
The case of Platynaspis luteorubra
Larvae of P. luteorubra occur with various ants tending
aphids, including L. niger,Myrmica spp. and Te-
tramorium caespitum, in both underground galleries
and on plants (Pontin 1959;Vo
¨lkl 1995; Godeau 2000).
The larvae and pupae have shapes unlike those of most
ladybirds, but similar to other myrmecophilous larvae,
such as those of some lycaenid butterflies and Micr-
odon hoverflies.
Vo
¨lkl (1995) has shown that P. luteorubra is a true
myrmecophile, and larvae are frequently found in
association with ant-tended aphids. Intensive field
studies on various plants showed that P. luteorubra
larvae occurred significantly more in ant-tended than
unattended colonies. The species has a range of
morphological and behavioural adaptations to life
with ants, thereby giving them access to ant-attended
resources. Ants do not recognise larvae of P. lute-
orubra as a threat to their attended aphids. This may
be due to the larva’s unusual coccid-like shape and
its slow inconspicuous movements. Vo
¨lkl also as-
sumed that the larvae produces ‘‘camouflage’’ chem-
icals. Studies in Germany have shown that when
larvae of P. luteorubra were moved between colonies
of the ant-tended aphids, A. fabae and Metopeurum
fuscoviride, the response of ants towards them
changed. Larvae moved to a new colony of conspe-
cific aphids were not attacked, while those moved to
a colony of the other species were (Oczenascheck
1997). Analysis of the larval cuticular lipids showed
that these were similar, both in type and quantity, to
those of their prey. As the cuticular lipids of these
two aphids differ both qualitatively and quantita-
tively, a change in prey led to a change in the
cuticular lipids and so the effectiveness of the larvae’s
chemical mimicry. This is a very efficient form of
scent mimicry because the larvae do not have to
manufacture different cocktails of mimetic chemicals
when feeding on different prey species.
Vo
¨lkl (1995) found that P. luteorubra pupae were
frequently attacked by L. niger, but were protected
from injury by dense long hairs. The chemical mimicry
of the larvae is not carried forward into the pupal
stage, probably because alcohols rather than lipids
dominate the cuticular compounds of pupae. In Bel-
gium, adult P. luteorubra frequently feed on A. fabae
tended by L. niger on Cirsium arvense (Godeau 2000).
Adults are often attacked by L. niger and respond ei-
ther by fleeing or by clamping down.
The myrmecophily of P. luteorubra is adaptive.
Larvae in ant-tended aphid colonies have higher prey
capture than those in unattended colonies, and adults
that develop in ant-tended colonies are larger than
those that develop in untended colonies (Vo
¨lkl 1995).
Habitat specialisation and the evolution
of myrmecophily in coccinellids
There are striking differences in the adaptations that
the two best-studied myrmecophilous coccinellids have
evolved to enable them to live with ants. Platynaspis
luteorubra larvae chemically mimic aphids, sequester-
ing mimetic chemicals from their prey. Coccinella
magnifica appears to use ant-repellent chemicals, as
well as physical and behavioural defences. Despite
these different adaptations, the main reason for myr-
mecophily is probably the same in the two species: to
enable them to feed on ant-tended aphids when other
aphids are scarce.
Comparative work on C. magnifica and C. septem-
punctata allows speculation on the evolution of myr-
mecophily. Sloggett and Majerus (2000a) showed that
C. septempunctata has some tolerance of F. rufa during
periods of aphid scarcity. Donisthorpe (1919–1920)
also wrote of C. 7-punctata ‘‘experimenting in a myr-
mecophilous existence’’. Furthermore, Bhatkar (1982)
observed large groups of this ladybird with F. polyc-
tena, and other Coccinella species (C. undecimpuncta-
ta,C. transversoguttata,C. trifasciata) have been
reported with ants, particularly in late summer (Brad-
ley and Hinks 1968; Bhatkar 1982). This suggests that
Coccinella species often facultatively coexist with ants.
Members of this genus are intolerant of low aphid
densities, and this may have driven them to evolve
some tolerance to ants when aphids are scarce
(Sloggett and Majerus 2000a). It seems feasible, then,
that the non-myrmecophilous ancestors of C. magnifica
occasionally had to prey upon ant-tended aphids, and
thus selection was imposed on them to evolve some
tolerance to ants. Additional selective advantages to
myrmecophily may have enhanced the behaviour over
time. These may have included more efficient use of
Popul Ecol (2007) 49:15–27 23
123
ant-tended prey species, reduced energetic costs asso-
ciated with prey switching (Hattingh and Samways
1992), reduced requirement for migrations, reduced
competition with other aphidophages and reduced
densities of ladybird predators and parasitoids (Slogg-
ett and Majerus 2000a).
Living in the forage range of aggressive ants may
give myrmecophilous ladybirds an advantage from
‘‘enemy-free space’’. Little work has been conducted
on the role of enemy-free space in the evolution of
habitat preferences in coccinellids, although it appears
to be weak when compared with food-related factors
(Sloggett and Majerus 2000b). In P. luteorubra, living
with ants greatly reduces the levels of infection by the
host-specific parasitoid wasp Homalotylus platynaspi-
dis (Vo
¨lkl 1995). Similarly, Majerus (1994) and Slogg-
ett et al. (2004) found lower levels of parasitism by
Dinocampus coccinellae in C. magnifica than in
C. septempunctata in the same habitat. However, in
captivity, in the absence of ants, D. coccinellae para-
sitism of C. magnifica is negligible. Thus, the low par-
asitism of C. magnifica by this wasp may be due to
factors intrinsic to the ladybird, possibly its highly
repellent defensive chemistry, rather than being
directly attributable to ants.
In research on habitat or host plant preferences, two
questions should be addressed. First, and obviously,
why does a species live in certain habitats or on certain
plants? The second question, which is often forgotten,
is why does a species not live in other habitats or on
other plants? Species with highly specialised ecologies,
such as myrmecophiles, are easiest to interpret.
In C. magnifica, we have some idea of why they live
with ants and not elsewhere. They do so to utilise a
food source protected from other predators by an
aggressive guardian, against whom they are themselves
well defended. But why does C. magnifica not live
anywhere else? Here we must speculate. Possibly
C. magnifica are poor competitors or lack efficient
defences against some predators and parasites. How-
ever, if this is so, more fundamental questions must
then be asked: why are they bad competitors and their
defences inadequate?
One possibility may follow from C. magnifica’s
specialisation to life with ants. Production of repellent
alkaloids is costly, reducing resources for other func-
tions, such as toxin production, or fighting ability. This
is a direct cost of immunity. A more indirect cost is that
when living with aggressive ants, the selection pres-
sures to maintain strong defences against a range of
predators and parasites are reduced, because ants ex-
clude these. If these ancestral defences have any cost,
the systems will be lost. In the early 1990s, Dr John
Barrett (personal communication) devised an inter-
esting analogy. If C. magnifica are the populace of the
United States, then the ants could be seen as a Rea-
ganesque Star Wars system, spreading a powerful
defensive umbrella over an area surrounding their
territory. Presence of this defensive system negates the
need for more conventional defences, and the costs of
these can be saved. One is safe as long as one stays
under the umbrella, but not if one strays.
Conclusion
Many predatory coccinellids encounter homopteran-
tending ants regularly because they both use resources
provided by Homoptera. Ants are thus important in
the ecology of many coccinellids. However, caution
should be taken when investigating the interactions
between coccinellids and ants. Perhaps too often the
reactions of coccinellids to ants have been viewed in
isolation. In reality, many of the defences used by
coccinellids when faced with aggressive ants are simply
modifications of general defences. So, it is those few
species of coccinellid that have the closest association
with ants, the myrmecophiles, that may be most illu-
minating. Here, the closeness of the association means
that many coccinellid behaviours to ants have evolved
specifically because of the association. The two best
studied myrmecophilous coccinellids, C. magnifica and
P. luteorubra, have already given insights into not only
specific inter-species interactions, but also the roles of
enemy free space, resource utilisation, inter-specific
competition and the evolution of habitat specificity.
Moreover, studies on these two species show both
illuminating similarities and differences. It is striking
that so little is known about ladybird–ant relationships,
with the paucity of knowledge about such interactions
in the tropics being most extreme. Certainly, other
myrmecophilous coccinellids await discovery. If other
ladybirds with close associations with ants are identi-
fied, then close scrutiny will surely provide novel in-
sights into a range of phenomena.
References
Addicott JF (1979) A multispecies aphid-ant association: density
dependence and species–specific effects. Can J Zool 57:558–
569
Agarwala BK Dixon AFG (1992) Laboratory study of cannibal-
ism and interspecific predation in ladybirds. Ecol Entomol
17:303–309
Attygalle AB, McCormick KD, Blankespoor CL, Eisner T,
Meinwald J (1993a) Azamacrolides: a family of alkaloids
from the pupal defensive secretion of a ladybird beetle
24 Popul Ecol (2007) 49:15–27
123
(Epilachna varivestis). Proc Natl Acad Sci USA 90:5204–
5208
Attygalle AB, Xu S-C, McCormick KD, Meinwald J (1993b)
Alkaloids of the Mexican bean beetle, Epilachna varivestis
(Coccinellidae). Tetrahedron 49:9333–9342
Bach CE (1991) Direct and indirect interactions between ants
(Pheidole megacephala), scales (Coccus viridis) and plants
(Pluchea indica). Oecologia 87:233–239
Banks CJ (1958) Effects of the ant Lasius niger (L.), on the
behaviour and reproduction of the black bean aphid, Aphis
fabae Scop. Bull Entomol Res 49:701–714
Banks CJ (1962) Effects of the ant Lasius niger (L.) on insects
preying on small populations of Aphis fabae (Scop.) on bean
plants. Ann Appl Biol 50:669–679
Banks CJ, Macaulay EDM (1967) Effects of Aphis fabae Scop.
And of its attendant ants and insect predators on yields of
field beans (Vicia faba L.). Ann Appl Biol 60:445–453
Bartlett BR (1961) The influence of ants upon parasites,
predators and scale insects. Ann Entomol Soc Am 54:543–
551
Berti N, Boulard M, Duverger C (1983) Fourmis et Coccinelles:
revue bibliographique et observations nouvelles. Bull Soc
Entomol Fr 88:271–275
Bhatkar AP (1982) Orientation and defense of ladybeetles
(Coleophera: Coccinellidae), following ant trail in search of
aphids. Folia Entomol Mex 53:75–85
Bradley GA (1973) Effect of Formica obscuripes (Hymenoptera:
Formicidae) on the predator-prey relationship between
Hyperaspis congressis (Coleoptera: Coccinellidae) and Tou-
meyella numismaticum (Homoptera: Coccidae). Can Ento-
mol 105:1113–1118
Bradley GA, Hinks JD (1968) Ants, aphids and jack pine in
Manitoba. Can Entomol 100:40–50
Brakefield PM (1985) Polymorphic Mu
¨llerian mimicry and
interactions with thermal melanism in ladybirds and a
soldier beetle: a hypothesis. Biol J Linn Soc 26:243–267
Bristow CM (1984) Differential benefits from ant attendance to
two species of Homoptera on New York Ironweed. J Anim
Ecol 53:715–726
Carroll CR, Janzen DH (1973) Ecology of foraging by ants.
Annu Rev Ecol Syst 4:231–257
Chapin EA (1966) A new species of myrmecophilous Coccinel-
lidae, with notes on the other Hyperaspini (Coleoptera).
Psyche 73:278–283
Corbara B, Dejean A, Cerdan P (1999) Une coccinelle myrme-
cophile associe
´ea
`la fourmi arboricole Dolichoderus bidens
(Dolichoderinae). Actes Coll Ins Soc 12:171–179
Daloze D, Braekman J-C, Pasteels JM (1995) Ladybird defence
alkaloids: structural, chemotaxonomic and biosynthetic
aspects (Col.: Coccinellidae). Chemoecology 5/6:173–183
DeBach P, Fleschner CA, Dietrick EJ (1951) A biological check
methoid for evaluating the effectiveness of entomophagous
insects. J Econ Entomol 44:763–766
Dechene R (1970) Studies of some behavioural patterns of
Iridomyrmex humilis Mayr (Formicidae, Dolichoderinae).
Wasmann J Biol 28:175–184
Disney RHL, Majerus MEN, Walpole M (1994) Phoridea
(Diptera) parasitising Coccinellidae (Coleoptera). Entomol-
ogist 113:28–42
Dixon AFG (1970) Quality and availability of food for a
sycamore aphid population. Symp Br Ecol Soc 10:271–287
Dixon AFG (1998) Aphid Ecology, 2nd edition. Blackie,
Glasgow
Dixon AGF (2000) Insect predator-prey dynamics: ladybird
beetles and biological control. Cambridge University Press,
Cambridge
Donisthorpe HStJK (1919–1920) The myrmecophilous ladybird
Coccinella distincta, Fald., its life history and association
with ants. Entomol Rec J Var 32:1–3
Eisner T, Eisner M (1992) Operation and defensive role of ‘‘gin
traps’’ in a coccinellid pupa (Cycloneda sanguinea). Psyche
99:265–274
Eisner T, Hicks K, Eisner M (1978) ‘‘Wolf-in-sheep’s-clothing’’
strategy of a predaceous insect larva. Science 199:790–794
Eisner T, Goetz M, Anaeshansley D, Ferstandig-Arnold G,
Meinwald J (1986) Defensive alkaloid in the blood of the
Mexican bean beetle (Epilachna varivestis). Experimentia
42:204–207
El-Ziady S (1960) Further effects of Lasius niger L. on Aphis
fabae Scopoli. Proc R Entomol Soc Lond A 35:30–38
El-Ziady S, Kennedy JS (1956) Beneficial effects of the common
garden ant Lasius niger L., on the black bean aphid, Aphis
fabae Scopoli. Proc R Entomol Soc Lond A 31:61–65
Godeau J-F (1997) Adaptations a
`la cohabitation avec des
fourmis: le cas de Coccinella magnifica Redtenbacher.
Me
´moire de D.E.A., Faculte
´des Sciences Agronomiques
de Gembloux, 42 pp
Godeau J-F (2000) Coccinelles amies des fourmis. 2/2/ Groupe
de Travail Coccinulla. Feuille de Contact 2:10–15
Godeau J-F, Hemptinne J-L, Verhaeghe J-C (2003) Ant trail: a
highway for Coccinella magnifica Redtenbacher (Coleop-
tera: Coccinellidae). In Soares AO, Ventura MA, Garcia V,
Hemptinne J-L (eds) Proceedings of the 8th International
Symposium on Ecology of Aphidophaga: Biology, Ecology
and Behaviour of Aphidophagous Insects. Arquipe
´lago:
Life and Marine Sciences, Supplement 5 pp 79–83
Gordon RD (1985) The Coccinellidae (Coleoptera) of America
North of Mexico. J NY Entomol Soc 93:1–912
Happ GM, Eisner T (1961) Hemorrage in a Coccinellid beetle
and its repellent effects on ants. Science 134:329–331
Harris RHTP (1921) A note on Ortalia pallens Muls. S Afr J Sci
17:170–171
Hattingh V, Samways MJ (1992) Prey choice and substitution in
Chilocorus spp. (Coleoptera: Coccinellidae). Bull Entomol
Res 82:327–334
Hays SB, Hays KL (1958) Food habits of Solenopsis saevissima
richteri Forel. J Ecol Entomol 52:455–457
Hemptinne J-L, Magro A, Majerus MEN (2005) Les Coccinelles.
Delachaux et Niestle
´, Paris
Hodek I (1996) Food Relationships. In: Hodek I, Honek A (eds)
Ecology of Coccinellidae. Kluwer, Dortrecht, pp 143–238
Ho
¨lldobler B, Wilson EO (1990) The ants. Springer, Berlin
Heidelberg New York
Holloway GJ, de Jong PW, Brakefield PM, de Vos H (1991)
Chemical defense in ladybird beetles (Coccinellidae). I.
Distribution of coccinelline and individual variation in
defence in 7-spot ladybirds (Coccinella septempunctata).
Chemoecology 2:7–14
Holloway GJ, de Jong PW, Ottenheim M (1993) The genetics
and cost of chemical defence in the 2-spot ladybird (Adalia
bipunctata L.). Evolution 47:1229–1239
Itioka T, Inoue T (1996) The role of predators and attendant ants
in the regulation of a population of the citrus mealybug
Pseudococcus citriculus in a Satsuma orange orchard. Appl
Entomol Zool 31:195–202
Jeffries MJ, Lawton JH (19849 Enemy free space and the
structure of ecological communities. Biol J Linn Soc 23:269–
286
Jiggins C, Majerus MEN, Gough U (1993) Ant defence of
colonies of Aphis fabae Scopoli (Hemiptera: Aphididae),
against predation by ladybirds. Br J Entomol Nat Hist
6:129–138
Popul Ecol (2007) 49:15–27 25
123
Jones TH, Blum MS (1983) Arthropod alkaloids: distribution,
functions, and chemistry. In: Pelletier SW (ed) Alkaloids
vol. 1, Chemical and biological perspectives. Wiley, New
York pp 33–84
de Jong PW, Holloway GJ, Brakefield PM, de Vos H (1991)
Chemical defence in ladybird beetles (Coccinellidae). II.
Amount of reflex fluid, the alkaloid adaline and individual
variation in defence in 2-spot ladybirds (Adalia bipunctata).
Chemoecology 2:15–19
Jutsum AR, Cherrett JM, Fisher M (1981) Interactions between
the fauna of citrus trees in Trinidad and the ants Atta
cephalotes and Azteca sp. J Appl Ecol 18:187–195
Kova
´r I (1996) Phylogeny. In: Hodek I, Honek A (eds) Ecology
of Coccinellidae. Kluwer, Dortrecht pp 19–31
Liepert C, Dettner K (1996) Role of cuticular hydrocarbons of
aphid parasitoids in their relationship to aphid-attending
ants. J Chem Ecol 22:695–707
MacKay WP (1983) Beetles associated with the harvester ants,
Pogonomyrmex montanus, P. subnitidus and P. rugosus
(Hymenoptera: Formicidae). Coleopt Bull 37:239–246
Mahdi T, Whittaker JB (1993) Do birch trees (Betula pendula)
grow better if foraged by wood ants? J Anim Ecol 62:101–116
Majerus MEN (1989) Coccinella magnifica (Redtenbacher)—a
myrmecophilous ladybird. Br J Entomol Nat Hist 2:97–106
Majerus MEN (1993) Notes on the inheritance of a scarce form of
the striped ladybird, Myzia oblongoguttata Linnaeus (Cole-
optera: Coccinellidae). Entomol Rec J Var 105:271–278
Majerus MEN (1994) Ladybirds.No. 81, New Naturalist Series.
Harper Collins, London
Majerus MEN, Majerus TMO (1997) Predation of ladybirds by
birds in the wild. Entomol Mon Mag 133:55–61
Mann WM (1911) On some northwestern ants and their guests.
Psyche 18:102–109
Mariau D, Julia JF (1977) Nouvelles reche
´rches sur la cochenille
du cocotier Aspidotus destructor (Sign). Ole
´agineux 32:217–
224
Marples NM (1993) Is the alkaloid in 2 spot ladybirds (Adalia
bipunctata) a defence against ant predation? Chemoecology
4:29–32
Marples NM, Brakefield PM, Cowie RJ (1989) Differences
between the 7-spot and 2-spot ladybird beetles (Coccinel-
lidae) in their toxic effects on a bird predator. Ecol Entomol
14:79–84
McLain DK (1980) Relationships among ants, aphids, and
coccinellids on wild lettuce. J Georgia Entomol Soc
15:417–418
Montgomery HWJr, Goodrich MA (2002) The Brachiacantha
(Coleoptera: Coccinelllidae) of Illinois. Trans Ill State Acad
Sci 95:111–130
Moore BP, Brown WV, Rothschild M (1990) Methylalkylpyra-
zines in apoosematic insects, their hostplants and mimics.
Chemoecology 1:43–51
Morgan CL (1896) Habit and instinct. London
Muggleton J (1978) Selection against the melanic morphs of
Adalia bipunctata (two-spot ladybird): a review and some
new data. Heredity 40:269–280
Oczenascheck C (1997) Chemische Ekologie der Entwicklungs-
stadien des Marienkaefers Platynaspis luteorubra Goeze.
Unpublished diploma thesis, University of Bayreuth
Orivel J, Servigne P, Cerdan Ph, Dejean A, Corbara B (2004)
The ladybird Thalassa saginata, an obligatory myrmecophile
of Dolichoderus bidens ant colonies. Naturwissenschaften
91:97–100
Pasteels JM, Deroe C, Tursch B, Brakeman JC, Daloze D,
Hootele C (1973) Distribution et activites des alcaloides
de
´fensifs des Coccinellidae. J Insect Physiol 19:1771–1784
Pontin AJ (1959) Some records of predators and parasites
adapted to attack aphids attended by ants. Entomol Mon
Mag 95:154–155
Pope RD (1979) Wax production by coccinellid larvae (Coleop-
tera). Syst Entomol 4:171–196
Proksch P, Wite L, Wray V, Hartmann T (1993) Ontogenic
variation of defensive alkaloids in the Mexican bean beetle
Epilachna varivestis (Coleoptera: Coccinellidae). Entomol
Gener 18:1–7
Radford P, Attygalle AB, Meinwald J, Smedley SR, Eisner T
(1997) Pyrrolidinooxazolidine alkaloids from two species of
ladybird beetles. J Nat Prod 60:755–759
Rathcke B, Hamrum CL, Glass AW (1967) Observations of the
interrelationships among ants and aphid predators. Mich
Entomol 1:169–173
Reimer NJ, Cope M-L, Yasuda G (1993) Interference of
Pheidole megacephala (Hymenoptera: Formicidae) with
biological control of Coccus viridis (Homoptera: Coccidae)
in coffee. Environ Entomol 22:483–488
Richards AM (1980) Defence adaptations and behaviour in
Scymnodes lividigaster (Coleoptera: Coccinellidae). J Zool
Lond 192:157–168
Richards AM (1985) Biology and defensive adaptations in
Rodatus major (Coleoptera: Coccinellidae) and its prey,
Monophlebus pilosior (Hemiptera: Margarodidae). J Zool
Lond A 205:287–295
Rosen D (1990) (ed) Armoured scale insects: their biology,
natural enemies and control. vol B Elsevier, Amsterdam
Rothschild M, Reichstein T (1976) Some problems associated
with the storage of cardiac glycosides by insects. Nova Acta
Leopoldina Supplement 7:507–550
Schro
¨der FC, Farmer JJ, Attygalle AB, Smedley SR, Eisner T,
Meinwald J (1998) Combinatorial chemistry in insects: a
library of defensive macrocyclic polyamines. Science
281:428–431
Seibert TF (1992) Mutualistic interactions of the aphid Lachnus
allegheniensis (Homoptera: Aphididae) and its tending ant
Formica obscuripes (Hymenoptera: Formicidae). Ann Ento-
mol Soc Am 85:173–178
Shi XW, Attygalle AB, Meinwald J (1997) Defense mechanisms
of arthropods. 149. Synthesis and absolute configuration of
two defensive alkaloids from the Mexican bean beetle,
Epilachna varivestis. Tetrahedron Lett 38:6479–6482
Silvestri F 1903 Contribuzioni alla conoscenza dei Mirmecophili,
I. Osservazioni su alcuni mirmecophili dei dintorni di
Portici. Ann Mus Zool R Univ Napoli 1:1–5
Sloggett JJ (1998) Interactions between coccinellids (Coleop-
tera) and ants (Hymenoptera: Formicidae), and the evolu-
tion of myrmecophily in Coccinella magnifica
Redtenbacher. Unpublished PhD thesis, University of
Cambridge
Sloggett JJ (2005) Are we studying too few taxa? Insights from
aphidophagous ladybird beetles (Coleoptera: Coccinelli-
dae). Eur J Entomol 102:391–398
Sloggett JJ, Majerus MEN (2000a) 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, Majerus MEN (2000b) Habitat preferences and diet
in the predatory Coccinellidae (Coleoptera): an evolution-
ary perspective. Biol J Linn Soc 70:63–88
Sloggett JJ, Majerus MEN (2003) Adaptations of Coccinella
magnifica, a myrmecophilous coccinellid to aggression
by wood ants (Formica rufa group). II. Larval behaviour,
and ladybird oviposition location. Eur J Entomol 100:337–
344
26 Popul Ecol (2007) 49:15–27
123
Sloggett JJ, Wood RA, Majerus MEN (1998) Adaptations 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
Sloggett JJ, Vo
¨lkl W, Schulze W, von der Schulenberg JH,
Majerus MEN (2002) The ant-associations and diet of the
ladybird Coccinella magnifica (Coleoptera: Coccinellidae).
Eur J Entomol 99:565–569
Sloggett JJ, Webberley KM, Majerus MEN (2004) Low parasit-
oid success on a myrmecophilous host is maintained in the
absence of ants. Ecol Entomol 29:123–127
Smith JB (1886) Ant’s nests and their inhabitants. Am Nat
20:679–687
Southwood TRE (1977) Ecological methods: with particular
reference to the study of insect populations, 2nd Edn.
Chapman & Hall, London
Sta
¨ger R (1929) Warum werden geweisse Insekten von den
Ameisen nicht verzehrt? Z Insbiol 24:227–230
Stephens DW, Krebs JR (1986) Foraging theory. Princeton
University Press, Princeton
Sterling WL, Jones D, Dean DA (1979) Failure of the red
imported fire ant too reduce entomophagous insect and
spider abundance in a cotton agroecosystem. Environ
Entomol 8:976–981
Strong DR, Lawton JH, Southwood TRE (1984) Insects on plants:
community patterns and mechanisms. Blackwell, Oxford
Tursch B, Braekman JC, Daloze D (1976) Arthropod alkaloids.
Experientia 32:401–407
Vinson SB, Scarborough TA (1989) Impact of the imported fire
ant on laboratory populations of cotton aphid (Aphis
gossypii) predators. Fla Entomol 72:107–111
Vo
¨lkl W (1995) Behavioural and morphological adaptations of
the coccinellid, Platynaspis luteorubra for exploiting ant-
attending resources (Coleoptera: Coccinellidae). J Insect
Behav 8:653–670
Vo
¨lkl W (1997) Interactions between ants and aphid parasitoids:
Patterns and consequences for resource utilization. Ecol
Stud 130:225–240
Vo
¨lkl W, Vohland K (1996) Wax covers in larvae of two
Scymnus species: do they enhance coccinellid larval sur-
vival? Oecologia 107:498–503
Wasmann E (1894) Kritisches Verzeichniss der Myrmekophilen
und Termitophilen Arthropoden. Verlag von Felix L.
Dames, Berlin
Wasmann E (1912) Neue Beitra
¨ge zur Kenntnis der Termito-
philen und Myrmecophilen. Zeits Wissens Zool 101:70–115
Way MJ (1954) Studies on the association of the ant Oecophylla
longinoda (Latr.) with the scale insect, Saissetia zanzibar-
ensis (Williams). Bull Entomol Res 45:113–134
Way MJ (1963) Mutualism between ants and honeydew produc-
ing Homoptera. Annu Rev Entomol 8:307–344
Wheeler WM (1911) An ant-nest coccinellid [Brachyancantha
quadripunctata (Mels.)]. J NY Entomol Soc 19:169–174
Wilson NL, Oliver AD (1969) Food habits of the imported fire
ant in pasture and pine forest areas in southeastern
Louisiana. J Econ Entomol 62:1268–1271
Witte L, Ehmke A, Hartmann T (1990) Interspecific flow of
alkaloids from plants via aphids to ladybirds. Naturwissens-
chaften 77:540–543
Popul Ecol (2007) 49:15–27 27
123
