ArticlePDF Available

Abstract and Figures

The epiphyte Dischidia major has highly modified leaves (‘pitchers’) that provide lodging for various ants, especially Philidris (Dolichoderinae) but also Cataulacus and Crematogaster (Myrmicinae). In return, the plants can obtain extra nutrients but this depends on intimate contact between the branching adventitious roots growing within inhabited pitchers and the organic debris brought in by the ants. D. major pitchers were sampled in two very different habitats in Thailand: coastal heathland in the south, and the canopy of a 30 m-high Dipterocarpus alatus in the north-East. We recognized only one species of Philidris in these two locations. Up to a few hundreds of workers and many brood were found in each pitcher. several dealate queens were all mated and egg-laying. Workers are polymorphic in size and morphometric analysis showed that large individuals have disproportionally big heads. Importantly, Philidris th01 divided each pitcher into compartments by building walls around the roots; this increases surface area for their brood but this selfish behaviour also matches the epiphyte’s trophic interests. The entrances of adjacent pitchers were often enclosed by soil runways, connecting separate pitchers into one extensive nest. This external accumulation of substrate may also benefit the epiphyte. Philidris th01 occurs in different habitats throughout Thailand, including disturbed vegetation. Some nests were found away from D. major, indicating that this ant-plant mutualism is not obligate.
Content may be subject to copyright.
ASIAN MYRMECOLOGY Volume 6, 49–61, 2014
ISSN 1985-1944 © Ch r i s t i a n Pe e t e r s a n d de C h a Wi W a t W i t a y a
Philidris ants living inside Dischidia epiphytes from Thailand
Ch r i s t i a n Pe e t e r s *1 a n d de C h a Wi W a t W i t a y a 2
1Laboratoire d’Ecologie, CNRS UMR 7625, Université Pierre et Marie Curie,
7 quai Saint Bernard, 75005 Paris, France
2Department of Forest Biology, Kasetsart University,
10900 Bangkok, Thailand
*Corresponding author's e-mail:
Abstract. The epiphyte Dischidia major has highly modied leaves (‘pitchers’)
that provide lodging for various ants, especially Philidris (Dolichoderinae)
but also Cataulacus and Crematogaster (Myrmicinae). In return, the plants
can obtain extra nutrients but this depends on intimate contact between the
branching adventitious roots growing within inhabited pitchers and the organic
debris brought in by the ants. D. major pitchers were sampled in two very
different habitats in Thailand: coastal heathland in the South, and the canopy of
a 30 m-high Dipterocarpus alatus in the North-East. We recognized only one
species of Philidris in these two locations. Up to a few hundreds of workers and
many brood were found in each pitcher. Several dealate queens were all mated
and egg-laying. Workers are polymorphic in size and morphometric analysis
showed that large individuals have disproportionally big heads. Importantly,
Philidris th01 divided each pitcher into compartments by building walls around
the roots; this increases surface area for their brood but this selsh behaviour
also matches the epiphyte’s trophic interests. The entrances of adjacent pitchers
were often enclosed by soil runways, connecting separate pitchers into one
extensive nest. This external accumulation of substrate may also benet the
epiphyte. Philidris th01 occurs in different habitats throughout Thailand,
including disturbed vegetation. Some nests were found away from D. major,
indicating that this ant-plant mutualism is not obligate.
Keywords: mutualism, symbiosis, myrmecophyte, ant-plant, domatia,
polygyny, worker polymorphism, Crematogaster
A striking testimony to the success of arboreal
ants is the evolution of intimate interactions with
plants. Many ant-plant mutualisms have been
described, with several degrees of sophistication
(Beattie 1985, Moog et al. 2003). Generally, the
ant partners are provided with housing and/or
food. In return, the plants are protected because
the ants deter herbivorous insects. Additional
trophic benets for plants concern only a
minority of mutualisms (e.g. Bazile et al. 2012).
This seems particularly important for epiphytes
that often face severe nutritional constraints. A
diversity of epiphytes are known to house ants,
and this privileged relationship was recognized
long ago (Bequaert 1922). Dischidia major
(Vahl) Merr. (junior synonym is D. rafesiana)
is a succulent creeper (Asclepiadaceae) with two
kinds of leaves growing off a central stem. Small
coin-like regular leaves contrast with pouched
leaves called ‘pitchers’ with the latter kind often
predominating (Fig. 1). Adventitious roots grow at
the leaf joints to attach the stem to the host plant,
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 49 26-May-14 2:37:32 PM
50 Christian Peeters and Decha Wiwatwitaya
Fig. 1. Cluster of pitcher leaves of Dischidia major growing on a tree trunk.
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 50 26-May-14 2:37:33 PM
Philidris living inside Dischidia epiphytes
Table 1. Demographics of Philidris th01 inhabiting ve clusters of Dischidia pitchers collected from separate
trees in Songkla province. All pitchers in a cluster were opened. Queens were all dealate. Many workers escaped
and are not included in the counts.
Pitchers No. queens No. workers No.
pupae No.
larvae eggs
A1 0 46 473 19 0
A2+3 1 444 647 139 #
A5 0 84 89 613 0
A7 0 75 372 69 0
cluster A (details above): 7 pitchers & soil runway; 5 inhabited and 2 vacant *
cluster B: 24 pitchers, no soil runway
3 pitchers inhabited: 1 queen, workers, pupae and larvae
3 pitchers vacant* but roots + soil. All others are completely empty
cluster C: 14 pitchers
7 pitchers inhabited: 4 had single queen + brood. A few males in 5 pitchers
6 pitchers vacant* but roots + soil; 1 small pitcher empty
cluster E: 14 pitchers
2 pitchers inhabited: 1 & 3 queens, few workers and brood (including eggs)
12 pitchers vacant* but roots + soil
cluster X: one pitcher contained 6 queens and sexual larvae
* vacant pitchers may contain a few workers but no brood
# not checked
but one root grows into the pitcher cavity through
an opening at the base. This root proliferates
inside pitchers that are inhabited by ants, usually
Philidris (subfamily Dolichoderinae). Most
pitchers contain ant-deposited debris, leading
Janzen (1974) and Huxley (1980) to predict
that the plants assimilate some of it. This was
conrmed by Treseder et al. (1995) who traced the
movement of carbon and nitrogen isotopes from
ants to plants; 29% of the host nitrogen originates
from ant debris. Moreover, stomata located
inside the pitchers can absorb ant-respired carbon
dioxide, providing 39% of the carbon in occupied
host plants. In general, the quality of ‘domatia’,
i.e. cavities produced by the plant to house ants,
is highly variable between plant species (Moog
et al. 2003); the pitchers of Dischidia major
must rank among the best because they are
completely enclosed with an easily defendable
small entrance.
In Thailand we sampled two populations
(850 km apart) of Philidris nesting in D. major.
We recognized only one species, referred to as
Philidris th01. In southern Thailand, Kaufmann
& Maschwitz (2006) studied Philidris spKfmA85
nesting in D. major, while P. myrmecodiae (=
Iridomyrmex myrmecodiae; Shattuck 1992,
1994) was studied in Sarawak (Janzen 1974) and
northern Thailand (Kerr 1912). Although this
mutualism is well described, we know little about
the colony and physical attributes of Philidris;
Tschinkel (2010) has emphasized that ‘bottom-
up’ data are essential to understand the ontogeny,
life history and evolution of ants. We counted and
measured the inhabitants of pitchers, and used
morphometrics to analyse worker polymorphism
and queen-worker differences. We dissected
queen ovaries and determined that colonies are
polygynous. We describe the partitions built by
Philidris inside the pitchers and discuss how
this self-interested behaviour maximizes trophic
benets for the host plant.
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 51 26-May-14 2:37:33 PM
52 Christian Peeters and Decha Wiwatwitaya
Fig. 2. Workers and dealate queen of Philidris th01 showing strong size dimorphism. Callows are very
lightly pigmented.
In January and November 2007, we collected
pitchers of Dischidia major in the Faculty of
Forestry Station, Kasetsart University, near
Sakaerat (Nakhon Ratchasima province).
Using binoculars, we checked for the presence
of Dischidia in the crown of a 30 m high
Dipterocarpus tree. An assistant climbed an
adjacent tree and dislodged the epiphytes using
a long bamboo pole. Clusters of pitchers thus
fell to the ground where we retrieved them in
one bag. Since the epiphytes were cut randomly,
we lack information on the spatial relationship
between clusters, as well as exact contents (ants
can move between pitchers during transit). A
few of these nests were maintained for months
in Paris, using plaster nests with a glass roof that
allowed behavioural observations. Temperature
(25°C) and the humidity of the plaster of Paris
nest were controlled. Ants were fed with crickets,
Bhatkar’s diet (Bhatkar & Whitcomb 1970) and
honeywater. A few queens were individually
marked with paint. When plants of D. major were
purchased from nurseries in Germany, the ants
readily moved in.
In December 2011 we collected in
two localities in Songkla province (Southern
Thailand): a forest in Ton-Nga-Chang (700 m
elevation, Hat Yai district) and a coastal heathland
in Cha-Na district. Here, D. major grows on
small trees at a height of 1 – 2 m. Accordingly we
could photograph and then isolate four clusters
of pitchers in separate plastic bags. To prevent
ant relocations among pitchers, these were
kept in a fridge (4°C) before being cut open for
examination of contents.
Variations in the shape of individuals
result from both size differences and allometry
(i.e. differential growth rates across body parts).
A random sample of 97 workers (one colony
from Sakaerat) was photographed and measured
(head width, posterior tibia length and thorax
dimensions) with Image J (
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 52 26-May-14 2:37:34 PM
Philidris living inside Dischidia epiphytes
Fig. 3. Measurements of 97 workers from the same colony of Philidris th01 in Sakaerat: (a) size frequency
distribution of head width, showing bimodality; (b) length of head relative to tibia. The slopes of the regression
lines correspond to the allometry coefcients: 1 for small workers (isometry;white points and dotted line) and
1.5 for large workers (allometry, black points and solid line). Slopes are signicantly different (SMA regression;
p = 0.005).
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 53 26-May-14 2:37:34 PM
54 Christian Peeters and Decha Wiwatwitaya
ij). Following Molet et al. (2007), we calculated
thorax volume as a mean of two volumes: lateral
area multiplied by dorsal width and dorsal area
multiplied by lateral height. To assess differences
between large and small individuals, we computed
growth rules between several traits (Peeters &
Molet 2010a). Thorax volume was also calculated
for queens (n = 12) to describe size dimorphism
relative to workers, as well as other species; such
measures are useful for comparative studies of
colony-founding ability in ants.
We dissected 12 dealate queens under
a stereomicroscope, and examined ovaries
and spermatheca. To elucidate the nature of
debris present inside the pitchers, the inner
partitions were examined by scanning electron
microscopy. Voucher specimens have been
deposited in Kasetsart Ant Museum (AMK) in
Bangkok, Australian National Insect Collection
(ANIC) in Canberra, and California Academy of
Sciences (CAS) in San Francisco, USA. Sakaerat
specimens in CAS that have been imaged are
CASENT0906672 (worker), CASENT0906673
(worker), CASENT0906674 (dealate queen);
Songkla specimens are CASENT0906669
(worker), CASENT0906670 (dealate queen),
CASENT0906671 (male) - see http://www.antweb.
Polygyny and worker polymorphism
We recognized only one morphospecies (Philidris
th01) in both locations. Pitchers of Dischidia
major collected in Songkla province contained up
to 444 Philidris workers as well as brood (Table
1). Some pitchers included mostly larvae, others
mostly pupae. Adult males and sexual larvae also
occurred. A maximum of six dealate queens were
collected together in the same pitcher.
Trophallaxis among workers, queens
and larvae was frequent. We never observed any
antagonism among queens. All dealated queens
were mated (n = 12), with about 20 ovarioles
in each ovary and many yolky oocytes. Their
spermatheca is kidney-shaped and it was always
full of sperm. As seen in Fig. 2, queens are
much bigger than workers, with large bulbous
compound eyes comprising 300 350 ommatidia,
compared with the smaller eyes of workers (50 –
60 ommatidia). Queens also have three prominent
ocelli on top of the head. Thorax volume of queens
was 28 times greater than that of workers.
For each of three separate traits (head
width, tibia length, thorax volume), the frequency
distribution of Philidris th01 workers (n = 97)
was bimodal (see Fig. 3a for head width). Growth
rules based on three pairs of measures (head-
tibia, head-thorax, thorax-tibia) indicated that
small individuals are isometric (body parts have
the same length ratios in different-sized workers),
whereas large individuals are allometric (body
parts have different ratios in different-sized
workers) with modied head shapes (Fig. 3b).
Our assessment is compatible with Shattuck’s
(1992) general diagnosis of Philidris workers:
“polymorphic, majors with ocelli (occasionally
monomorphic)”, except that we did not nd traces
of ocelli in large workers from Thailand.
Structure of ant nests inside pitchers
Many of the pitchers were inhabited (Table 1), and
ant presence was associated with both extensive
root development and a variable quantity of
debris (Fig. 4). Some pitchers had clearly never
been inhabited, with tiny or unbranched internal
roots. Other pitchers lacked workers and brood
even though there were highly branched roots
and debris (Table 1); the ants had probably
moved out of these to settle in adjacent pitchers.
Pitchers offer a large empty space, much of which
is unavailable to the ants unless they structure it.
Philidris build partitions by using the internal
roots as framework (Fig. 5). They bring debris
from outside as construction material. The strong
cementing of the partitions allowed two tiny
strips to be examined by SEM, although this was
not very conclusive, showing mostly plant tissues
and no pieces of insect exoskeleton (Fig. 6).
In several clusters, soil runways had been
built between the bases of adjacent pitchers (Fig.
7). Runways built by ants looked distinct from
the more solid tubes built by termites (usually
mud mixed with saliva and faeces).
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 54 26-May-14 2:37:34 PM
Philidris living inside Dischidia epiphytes
Fig. 4. Vacant and inhabited pitchers of D. major, cut away to reveal presence or absence of roots and debris.
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 55 26-May-14 2:37:36 PM
56 Christian Peeters and Decha Wiwatwitaya
Other inhabitants of the pitchers
Both in Ton-Nga-Chang and Cha-Na, some
pitchers were inhabited by Crematogaster
rogenhoferi (workers, larvae and pupae). Plant
bres were woven together around the internal
adventitious roots, and this construction was very
distinct from Philidris partitions. Crematogaster
was equally able to build runways between
pitchers by weaving bres. There were large
carton nests of C. rogenhoferi on neighbouring
trees (away from pitchers), hence it is possible
that the colonies in pitchers were newly founded.
Indeed, one dealate queen and many eggs were
collected in one of four adjacent pitchers, which is
clear evidence of an incipient colony. Cataulacus
granulatus was also found in some pitchers (few
workers, larvae and winged gynes), but the root
was short and unbranched, and there was no
debris. In Sakaerat, several pitchers growing in
the canopy of Dipterocarpus were inhabited by
Dolichoderus thoracicus.
In Philidris colonies from Sakaerat,
many workers carried one large reddish oribatid
mite (Order Oribatulidae) attached on the tarsi of
one hind leg (sometimes both legs). In Songkla,
the same mite was also riding on the ants’ tarsi,
and some were found hiding in the debris in
pitchers (including vacant ones). We found no
other myrmecophiles, no scale insects on the
inner leaf surfaces and no obvious fungi growing
on debris.
Distribution in Thailand
Philidris is widely distributed across different
bioclimatic regions in Thailand, and it often
occurs in secondary forests, including rubber
plantations. Examination of specimens deposited
in the Kasetsart Ant Museum indicates the
following provinces of origin: Kanchanaburi
(mixed deciduous forest and teak plantation),
Narathiwat (along trail in rainforest), Pattani,
Chumphon (mangrove and rubber plantation),
Kanchanaburi (mixed deciduous forest and
disturbed vegetation), Chiang Mai (Doi Ang
Khang 1300m elevation, roadside and plantation,
Doi Chiang Dow 300 m mixed deciduous
forest), Chanthaburi (along trail in rainforest),
Nakhonratchasima (Khao Yai NP, edge of hill
evergreen forest), Trad (mangrove forest). Despite
extensive size polymorphism among nestmate
workers, Philidris is morphologically invariant in
Thailand and appears to be just one species. In the
South, Kaufmann (2002) reported one species of
Philidris (spKfmA85) nesting in D. major, but we
have been unable to examine these specimens.
In Sakaerat, some Philidris nests were
found away from Dischidia, in the rotting bark of
small trees (3 – 5m high), or in decaying branches
or stumps close to ground. Free-living Philidris
construct their shelters in a manner similar to
pitcher-inhabiting colonies. Kaufmann (2002)
described extensive carton nests of Philidris
spKfmA37 on stems or branches.
Reproductive strategy
Established colonies of Philidris th01 are clearly
polygynous, but we speculate that they start off
with a single foundress (see Peeters & Molet
(2010b) for a review of strategies of colonial
reproduction). The presence of inhabited
Dischidia pitchers isolated in the 30 m high
crown of Dipterocarpus attests to the ying
and searching ability of Philidris foundresses.
Independent foundation in ants is difcult to study
in the eld (small time window during the year),
but aspects of queen morphology indicate that
foundressess have sufcient metabolic reserves
to raise the rst worker brood without foraging
outside (i.e. claustral foundation): (1) overall size
dimorphism relative to workers is striking (Fig.
2), and the large difference in thorax volume
(28x) indicates the presence of large wing muscles
that can be resorbed to feed the rst larvae; (2)
the rst thorax segment (pronotum) is small,
revealing that neck muscles are much reduced,
hence queens do not forage (Keller et al. 2014).
A foundress that succeeds in nding an empty
pitcher can presumably produce her rst workers
in relative safety. Given high queen fecundity
(many ovarioles), a colony can quickly grow in
size and expand into adjacent pitchers. In Sarawak,
Janzen (1974) reported each of 14 colonies of P.
myrmecodiae to have a single physogastric (i.e.
abdomen is stretched out to accommodate highly
developed ovaries) queen, and these may have
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 56 26-May-14 2:37:36 PM
Philidris living inside Dischidia epiphytes
Fig. 5. Inner partitions built by Philidris th01 ants with debris, using roots as a frame.
Fig. 6. Scanning electron micrograph of a partition built inside a Dischidia pitcher by Philidris. Plant material is
criss-crossed by the epiphyte roots.
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 57 26-May-14 2:37:37 PM
58 Christian Peeters and Decha Wiwatwitaya
been the original foundresses. In Philidris th01,
secondary polygyny could result from accepting
young queens, possibly daughters of the original
foundress that mate close to the nest. In the
obligate symbiosis between Crematogaster and
Macaranga, Feldhaar et al. (2000) interpreted
secondary polygyny to be an adaptation to
extend the life span of colonies, enabling them
to use resources of the host plant continuously.
Kaufmann (2002) described monogyny in
Philidris spKfmA37. Another species of Philidris
nesting in a different epiphyte (Hydnophytum
in Papua New Guinea) was reported to be
monogynous (Maeyama & Matsumoto 2000). If
polygyny is indeed secondary in Philidris th01,
future studies must attempt to distinguish between
new and old colonies.
Benets of ant-plant mutualism
We documented that inhabited pitchers are often
packed with workers and brood. The quantity
of larvae and pupae was particularly striking,
lling up all vacant spaces among the network
of roots and partitions. Pitchers become spatially
divided into many compartments as a result of
the ants’ building behaviour, and this allows
for optimal use of the entire volume. Debris
are brought from outside to build walls, which
causes extensive root growth and branching,
and in turn more roots may encourage the ants
to continue building. Roots are used as a frame
for the partitions, thus creating intimate contact
appropriate for absorption. Tiny root hairs were
seen developing wherever roots entered in
contact with any substrate (see also Janzen 1974).
Kerr’s (1912) observations of “clay mixed with
bits of wood and other vegetable matter” are not
contradicted by our SEM data. Philidris scavenge
on dead or dying arthropods, and also obtain
honeydew from various homopterans outside the
pitchers (Kaufmann & Maschwitz 2006). Thus it
is not surprising that insect exoskeletons can be
present in the ants’ debris, even though we found
very few, unlike Janzen (1974) who stated that
P. myrmecodiae differs in behaviour from other
arboreal ants that throw their refuse out of the
entrance. The benets for the two partners in this
mutualism are clear: ants get a safe and spacious
home that is easily defended (one small entrance)
against other ants. The epiphyte is able to use
ant-deposited debris, in addition to faeces and
discarded food, as a nitrogen source, although
this trophic benet depends on the self-interested
building activity of the ants. In contrast, other
myrmecophytes that chase herbivorous insects
or remove encroaching vegetation behave
specically to help the plant.
In addition to internal partitions, the ants
also build soil runways that connect the entrances
of neighbouring pitchers, which can sometimes
be extensive (Fig. 7). This accumulation of
substrate is also likely to benet the epiphyte,
e.g. short-term water storage for the adventitious
roots clinging to the stem. Runways seem to
be a general characteristic of Philidris nests,
including species that are not associated with D.
major (e.g. Kaufmann 2002). Runways strongly
suggest frequent movement of workers and brood
between pitchers. Accordingly it is semantic
whether a nest corresponds to a single pitcher
or one cluster of pitchers, and ‘polydomy’ must
be used with caution. We assume that adjacent
pitchers are all part of the same colony but we did
not investigate whether there can be other clusters
belonging to the same colony on the same tree.
Total colony size remained unknown. Kaufmann
(2002) states that a colony of Philidris spKfmA37
can reach 18000 workers (based on an estimate of
7 workers/cm3, obtained from counting workers
and brood in six nests).
Asclepiads generally contain poisonous
latex which protects against most herbivores.
Although Janzen (1974) dismissed protection
by the ants, we observed that upon the slightest
disturbance a large number of ants streamed out
with open mandibles (Fig. 7). A large patrolling
force seems advantageous to defend the plant
host as well as to gather resources. Moreover, the
large workers with proportionally larger heads
(i.e. more powerful mandible muscles) can be
especially efcient.
Pitchers inhabited by Crematogaster
ants lacked accumulations of debris. Their
partitions are built with vegetable carton, and it
is unknown if they can be exploited by the plant
roots. It is possible that different ant species
compete over pitchers. Beccari (1884) found D.
major in Java to be inhabited by Dolichoderus
thoracicus (=bituberculatus) or Crematogaster
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 58 26-May-14 2:37:37 PM
Philidris living inside Dischidia epiphytes
Fig. 7. Soil runways built by the ants to connect the entrances of adjacent pitchers. The central stem of Dischidia
can no longer be seen.
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 59 26-May-14 2:37:38 PM
60 Christian Peeters and Decha Wiwatwitaya
brevis. Similar to our observations in Songkla,
Kerr (1912) collected Cataulacus granulatus
inside Dischidia pitchers in northern Thailand.
Philidris th01 is an opportunist
Dischidia major appears widely distributed across
SE Asia (Rintz 1980, Weir & Kiew 1986). We
collected it in strikingly different microhabitats
in Thailand, coastal heathland and high canopy,
both exposed locations. Generally, a limiting
factor on epiphyte growth is nutrient deciency.
It is possible that mutualism with ants gives
Dischidia a competitive advantage over other
epiphytes in nutrient-poor microhabitats. In other
Dischidia species that lack pitcher leaves, some
roots penetrate Crematogaster nests (in cavities
inside living branches) by following tunnels used
by ants, and ‘scavenge’ their waste material (Weir
& Kiew 1986).
Philidris th01 is distributed throughout
Thailand (except the drier eastern part),
including disturbed habitats and plantations. Its
opportunistic nature is conrmed by the presence
of occasional nests in decaying wood or bark near
the ground, i.e. away from Dischidia. In Sarawak,
P. myrmecodiae also nests in other epiphytes
(Hydnophytum and Myrmecodia; Janzen 1974).
Kaufmann (2002) studied other species of
Philidris that build carton nests connected by
runways, away from Dischidia.
Many ant-plants provide trophic rewards
(specialized food bodies, or extraoral nectar)
for ants. Philidris th01 does not depend on its
epiphyte for food, and this may explain the lack
of an obligate relationship. Its nesting behaviour
resembles that of congeneric species not involved
with ant-house epiphytes. Similarly, comparisons
with related Dischidia species are useful to
understand the evolution of this well-matched
ant-plant mutualism.
Charlotte Holgate (Erasmus intern) did laboratory
observations and measurements during 2008.
Mathieu Molet analysed the morphometric data.
We thank Sirisak Jodnok for help in Sakaerat,
Chakkrapath Dulyaphat for organizing eldwork
in Songkla province, Patrick Landmann for the
photograph in Fig. 2, Nestor Fernandez for mite
identication, and Sasitorn Hasin for checking
specimens of Philidris. Mathieu Molet and Thibaud
Monnin gave useful comments on the manuscript,
and so did anonymous referees. Thanks are also
due to Tom Fayle for language editing.
Bhatkar A and Whitcomb WH, 1970. Articial
diet for rearing various species of ants. The
Florida Entomologist 53(4): 229 – 232.
Bazile V, Moran JA, Le Moguédec G, Marshall
DJ and Gaume L, 2012. A carnivorous
plant fed by its ant symbiont: A unique
multi-faceted nutritional mutualism. PLoS
ONE 7(5): e36179.
Beattie AJ, 1985. The evolutionary ecology of ant-
plant mutualism. Cambridge University
Press, Cambridge UK, 175 p.
Beccari O, 1884. Piante ospitatrici ossia piante
formicarie della Malesia e della Papuasia.
Malesia(Genoa) 2: 1 – 340.
Bequaert J, 1922. Ants in their diverse relations to
the plant world. Bulletin American Museum
of Natural History 45: 333 – 583.
Feldhaar H, Fiala B, Rosli H and Maschwitz U,
2000. Maintaining an ant-plant symbiosis:
secondary polygyny in the Macaranga
triloba - Crematogaster sp. association.
Naturwissenschaften 87: 408 – 411.
Huxley CR, 1980. Symbiosis between ants and
epiphytes. Biological Reviews 55: 321 – 340.
Janzen DH, 1974. Epiphytic myrmecophytes in
Sarawak: mutualism through the feeding
of plants by ants. Biotropica 6: 237 – 259.
Kaufmann E, 2002. Southeast Asian Ant-Gardens-
Diversity, ecology, ecosystematic
signicance, and evolution of mutualistic
ant-epiphyte associations. PhD Thesis,
Johann Wolfgang Goethe-Universität,
Frankfurt am Main. Downloaded
from publikationen.ub.uni-frankfurt.
de/les/5337/kaufmannEva.pdf on 25
November 2013.
Kaufmann E and Maschwitz U, 2006. Ant-
gardens of tropical Asian rainforests.
Naturwissenschaften 93: 216 – 227.
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 60 26-May-14 2:37:39 PM
Philidris living inside Dischidia epiphytes
Keller RA, Peeters C and Beldade P, 2014.
Evolution of thorax architecture in ant
castes highlights trade-off between ight
and ground behaviors. eLIFE 3: e01539.
Kerr AFG, 1912. Notes on Dischidia rafesiana
Wall. and Dischidia nummularia, Blume.
ScienticProceedings of the Royal Dublin
Society13: 292 – 315.
Maeyama Tand Matsumoto T, 2000. Colonial
system of Philidris ants (Formicidae;
Dolichoderinae) occupying epiphytic
myrmecophytes in a tropical mangrove
forest. Tropical Ecology 41: 209 – 216.
Molet M, Peeters C and Fisher BL, 2007.
Winged queens replaced by reproductives
smaller than workers in Mystrium ants.
Naturwissenschaften 94: 280 – 287
Moog J, Fiala B, Werner M, Weissog A, Saw
L Gand, Maschwitz U, 2003. Ant-plant
diversity in Peninsular Malaysia, with
special reference to the Pasoh Forest
Reserve. In: Pasoh: ecology of a lowland
rain forest in southeast Asia (Okuda T,
Manokaran N, Matsumoto Y, Niiyama K,
Thomas SC, Ashton PS, eds), Springer
Press, Tokyo, 459 – 494.
Peeters C and Molet M, 2010a. Evolution of
advanced social traits in phylogenetically
basal ants: striking worker polymorphism
and large queens in Amblyopone australis.
Insectes Sociaux 57: 177 – 183.
Peeters C and Molet M, 2010b. Colonial
reproduction and life histories. In: Ant
Ecology (Lach L, Parr C, Abbott K, eds),
Oxford University Press, 159 – 176.
Rintz RE, 1980. The Peninsular Malayan species
of Dischidia (Asclepiadaceae). Blumea
26: 81 – 126.
Shattuck SO, 1992. Review of the dolichoderine
ant genus Iridomyrmex Mayr with
descriptions of three new genera
(Hymenoptera: Formicidae). Journal of
the Australian Entomological Society 31:
13 – 18.
Shattuck SO, 1994. Taxonomic catalog
of the ant subfamilies Aneuretinae
and Dolichoderinae (Hymenoptera:
Formicidae). University of California
Publications in Entomology 112: 1 – 241.
Treseder KK, Davidson DW and Ehleringer
JR, 1995. Absorption of ant-provided
carbon dioxide and nitrogen by a tropical
epiphyte. Nature 375: 137 – 139.
Tschinkel W, 2010. Back to basics: sociometry and
sociogenesis of ant societies (Hymenoptera:
Formicidae). Myrmecological News 14:
49 – 54.
Weir JS and Kiew R, 1986. A reassessment of
the relations in Malaysia between ants
(Crematogaster) on trees (Leptospermum
and Dacrydium) and epiphytes of the
genus Dischidia (Asclepiadaceae)
including ‘ant-plants’. Biological Journal
of the Linnean Society 27: 113 – 132.
A Journal of the International Network for the Study of Asian Ants
Communicating Editors: Himender Bharti & Tom M. Fayle
4 - AM 6 - Philidris ants living inside Dischidia epiphytes.indd 61 26-May-14 2:37:39 PM
... Philidris is a genus of arboreal ants that build carton nests and runway galleries (Kaufmann 2002). When colonizing Dischidia pitcher leaves, they optimize space use by building carton platforms, using adventitious roots as a frame (Kerr 1912;Peeters and Wiwatwitaya 2014). They also accumulate debris within the leaf, eventually filling the domatium (Janzen 1974). ...
... They also accumulate debris within the leaf, eventually filling the domatium (Janzen 1974). Contact with carton or debris induces adventitious roots to produce fine root hairs, suggesting that the ants give their host plant access to nutrients in ant-generated materials (Janzen 1974;Peeters and Wiwatwitaya 2014). Nitrogen-deprived plants such as epiphytes would greatly benefit from such a strategy (Janzen 1974;Thompson 1981). ...
... Our investigation provides new information on microbial associates of the symbiosis between D. major and Philidris ants, complementing the previous studies on this system (e.g. Kerr 1912;Janzen 1974;Treseder et al. 1995;Kaufmann and Maschwitz 2006;Peeters and Wiwatwitaya 2014), which focused primarily on the interaction between ants and their host plant. We show that pitcher leaves occupied by ants host various species of chaetothyrialean and capnodialean fungi, a phenomenon known in most ant-plant symbioses investigated to date ). ...
Full-text available
Dischidia major is an epiphyte with pitcher leaves that serve as nests for ants. We investigated this ant-plant symbiosis in two sites in southeastern Thailand, Rayong and Trat, using a morphological and molecular approach. In our study sites, D. major was colonized by one monomorphic ant species of genus Philidris. The inner surface of the pitcher leaves had a black and green lining composed of intermingled coccoid cells and filaments of algae and fungi, reminiscent of a biofilm structure. Microscopic investigation of the algae suggested they belonged to Trebouxia (coccoid cells) and Trentepohliaceae (filaments). Molecular investigation of environmental samples and pure cultures of the fungi revealed five species of Chaetothyriales and four species of Capnodiales, among which two have already been isolated from ant-plant symbioses in Africa and South America and five were described species known from various environments around world. One appears to be an undescribed species. Thus, most fungal associates were likely ubiquitous species. Our study highlights the need to include the identity and functional ecology of microbes in studies of the evolutionary and functional ecology of ant-plant symbioses.
... For example, leaves in some epiphytic species of Dischidia and Hoya are modified into "pitchers" that are inhabited by ants. These plants offer room for the ants to raise their young and in return, the ants bring nutrients through their debris and confer protection to the plant (Janzen, 1974;Peeters and Wiwatwitaya, 2014). ...
... For example, leaves in some epiphytic species of Dischidia and Hoya are modified into "pitchers" that are inhabited by ants. These plants offer room for the ants to raise their young and in return, the ants bring nutrients through their debris and confer protection to the plant (Janzen, 1974;Peeters and Wiwatwitaya, 2014). ...
Full-text available
Apocynaceae (the dogbane and milkweed family) is one of the ten largest flowering plant families, with approximately 5,350 species and diverse morphology and ecology, ranging from large trees and lianas that are emblematic of tropical rainforests, to herbs in temperate grasslands, to succulents in dry, open landscapes, and to vines in a wide variety of habitats. Despite a specialized and conservative basic floral architecture, Apocynaceae are hyperdiverse in flower size, corolla shape, and especially derived floral morphological features. These are mainly associated with the development of corolline and/or staminal coronas and a spectrum of integration of floral structures culminating with the formation of a gynostegium and pollinaria-specialized pollen dispersal units. To date, no detailed analysis has been conducted to estimate the origin and diversification of this lineage in space and time. Here, we use the most comprehensive time-calibrated phylogeny of Apocynaceae, which includes approximately 20% of the species covering all major lineages, and information on species number and distributions obtained from the most up-to-date monograph of the family to investigate the biogeographical history of the lineage and its diversification dynamics. South America, Africa, and Southeast Asia (potentially including Oceania), were recovered as the most likely ancestral area of extant Apocynaceae diversity; this tropical climatic belt in the equatorial region retained the oldest extant lineages and these three tropical regions likely represent museums of the family. Africa was confirmed as the cradle of pollinia-bearing lineages and the main source of Apocynaceae intercontinental dispersals. We detected 12 shifts toward accelerated species diversification, of which 11 were in the APSA clade (apocynoids, Periplocoideae, Secamonoideae, and Asclepiadoideae), eight of these in the pollinia-bearing lineages and six within Asclepiadoideae. Wind-dispersed comose seeds, climbing growth form, and pollinia appeared sequentially within the APSA clade and probably work synergistically in the occupation of drier and cooler habitats. Overall, we hypothesize that temporal patterns in diversification of Apocynaceae was mainly shaped by a sequence of morphological innovations that conferred higher capacity to disperse and establish in seasonal, unstable, and open habitats, which have expanded since the Eocene-Oligocene climate transition.
... The epiphyte is now ready to absorb the antdeposited debris, in addition to faeces and discarded food, as a nitrogen source. Moreover, stomata located inside the pitchers can also absorb ant-respired carbon dioxide (Peeters and Wiwatwitaya, 2014). ...
... Some species are myrmecophytes, plants that have a symbiotic relationship with ants (Rintz, 1980;Livshultz et al., 2005). Myrmecophilous Dischidia host ants in two ways: some species [e.g., D. major (Vahl) Merr.] have pitcher-shaped leaves in which ants live and in which the plant sends its roots (Janzen, 1974;Huxley, 1980;Peeters and Wiwatwitaya, 2014), while other species have convex, shingle-leaves under which ants live and plant roots proliferate. Only the latter group, the shingle-leaf climbers (Table 1), is considered here. ...
The curious habit of shingle-leaf climbers – root-climbing plants whose leaves are closely adpressed to the phorophyte and often overlap like shingles – has attracted the attention of both botanists and horticulturists for more than a century. The habit has arisen in ten families, 22 genera, and at least 158 species and is especially common in several genera of Araceae and Marcgravia (Marcgraviaceae). Herein, the species are tabulated, and various hypotheses for the evolution of the habit are reviewed. Two hypotheses that emerge as having explanatory power for understory shingle-leaf climbers are 1) Trapping & Recycling CO2 and 2) Stemflow Nutrient Capture, but other hypotheses may also have support in some cases. Three hypotheses (Balancing Carbon Allocation, Avoiding Damage from Falling Objects, and Avoiding Herbivory) have some support for some species. One hypothesis (Protecting against Desiccation of Roots and Leaves) has some support for shingle-leaf climbers in exposed, sunny habitats (viz., Hoya, Dischidia, Cattleya cernua and other orchids). Different selective pressures may have led to convergence on the shingle-leaf habit in different habitats. Moreover, these hypotheses are not necessarily mutually exclusive. Few hypotheses have been explicitly tested, and so the adaptive significance of the shingle-leaf climber habit remains uncertain.
... Another notable ant genus occurring frequently as prey among both upper and lower pitchers is Philidris (Table 4). Little is known about the biology of this genus, but some species have been observed to be closely associated with myrmecophytes, such as the epiphyte Dischidia major (Peeters & Wiwatwitaya, 2014). A similar mutualistic association may exist between N. rafflesiana and Table 5. Mean number of each prey species trapped per pitcher in which the prey taxon was found. ...
Full-text available
The Raffles' pitcher plant (Nepenthes rafflesiana Jack, Nepenthaceae) is a carnivorous plant found naturally across multiple habitat types both within and outside of the Central Catchment Nature Reserve (CCNR)-the largest contiguous forested habitat in Singapore. Nepenthes rafflesiana produces lower and upper pitchers during distinct ontogenetic stages, and both of these pitcher types trap highly diverse but compositionally differentiated spectra of invertebrate prey. The fluid-filled pitchers of N. rafflesiana are also habitats for many specialised aquatic invertebrate species, known as inquilines. However, neither the prey spectra nor inquiline communities of N. rafflesiana pitchers are well characterised to a high taxonomic resolution. In this study, we surveyed pitcher-prey assemblages and inquiline communities in 78 N. rafflesiana pitchers within and outside of the CCNR in Singapore, as well as the plant communities in the 10 × 10 m plots in which the N. rafflesiana plants were documented. Plant communities of N. rafflesiana plots within the CCNR were found to be much richer in species and differed compositionally from those outside of the CCNR. Inquiline communities, too, were observed to differ compositionally between pitchers within and outside of the CCNR, as well as between pitcher types. However, asymptotic inquiline species richness was not significantly lower in pitchers outside of the CCNR, as compared to those within it. Lower pitchers generally contained more inquiline species than upper pitchers. Finally, prey assemblages differ in composition only marginally between pitchers within and outside of the CCNR, but were strongly differentiated between pitcher types.
... The intricate and mutually beneficial associations existing between ants and tropical forest plants were first described for Asia by Beccari (1884Beccari ( -1886 and elaborated upon by Van Leeuwen (1913, 1923a. Subsequently an extensive body of literature has been generated for tropical Asia, notably for Euphorbiaceae (Macaranga -see for example Fiala et al. 1991), Rubiaceae (Huxley 1978;Razafimandimbison et al. 2005;Jebb & Huxley 2019), Melastomataceae (Clausing 1997), Apocynaceae (Kleijn and van Donkelaar 2001;Peeters & Wiwatwitaya 2014;Weissflog et al. 2017), and for the palm genus Korthalsia (Chan et al. 2012;Miler et al. 2016). Good general overviews for one lowland area of Peninsular Malaysia are provided of Fiala and Saw (2003) and Moog et al. 2003. ...
Full-text available
The presence of stipular and leaf blade extra floral nectaries and associated ant activity, including brood raising within stipules, is reported for saplings of Shorea macrophylla [sect. Pachycarpae] in Kuching Division, Sarawak.
Full-text available
Thailand has a great diversity of ant fauna as a zoogeographical crossroads and biodiversity hotspot. The last publication presenting a Thai ant checklist was published in 2005. In the present paper, based on an examination of museum specimens and published records, a comprehensive and critical species list of Thai ants is synthesized. Currently, 529 valid species and subspecies in 109 genera among ten subfamilies are known from Thailand with their diversity and distribution within 77 provinces presented and assigned to six geographical regions. Furthermore, Thailand is the type locality for 81 ant species. Forty-one species are here newly recorded for Thailand with photographs illustrating these species. The checklist provides information on distribution and a comprehensive bibliography. This study will also serve as a guide for the upper northeast and central Thailand, which are poorly sampled; a comprehensive reference list relating to endemic taxa and localities where conservation is an important priority, thus an essential resource for policy makers and conservation planners concerned with the management of insect diversity in Thailand; and a list of exotic ant species found in Thailand, which could possibly impact the ecological balance.
Full-text available
Two new species of Nepenthes from Terengganu, Peninsular Malaysia, N. latiffiana M. N. Faizal, A. Amin & N. Dome and N. domei M. N. Faizal, A. Amin & A. Latiff, are described and illustrated.
Woody climbers, vines, perennial herbs, trees or shrubs, more rarely annuals, sometimes with large water-storing tubers or a xylopod, sometimes succulent, with large grappling hooks and/or tendrils in several lianoid genera of Willughbeieae; latex in non-articulated laticifers present, most commonly white, but in some genera usually translucent and in others yellowish or reddish. Leaves simple and usually entire, very rarely dentate or repand, usually isophyllous, but often anisophyllous in Tabernaemontaneae-Tabernaemontiinae, sometimes with distinctly different juvenile and adult foliage, normally petiolate, sometimes sessile, usually opposite, less frequently alternate or whorled (whorled phyllotaxis characteristic for a number of Rauvolfioid genera); stipules usually absent or small and caducous, sometimes enlarged and fused into dentate interpetiolar collars (a few Periplocoid genera), commonly with interpetiolar lines or ridges, sometimes the petioles of a leaf pair connate at the node, forming a short ocrea, which may be expanded into small intrapetiolar flaps clasping the stem (Tabernaemontaneae), almost always with colleters in the axil of the leaf, sometimes on the petiole, in a cluster adaxially at the juncture of petiole and lamina or along the midrib above, occasionally with abaxial domatia in the axils of the secondary veins (mainly in Apocynoids). Flowers perfect, rarely functionally dioecious, often scented, sessile or more commonly pedicellate, in solitary or more commonly in axillary, extra-axillary or terminal multi-flowered cymes, panicles or thyrses, sometimes appearing as an axillary fascicle. Perianth almost always actinomorphic, very rarely slightly zygomorphic; calyx almost always 5- (rarely 4- or 6–7-)merous, lobes normally quincuncially arranged, synsepalous or aposepalous, commonly with colleters, in Periplocoideae, Secamonoideae and Asclepiadoideae these are usually in the sinuses, but in some Rauvolfioids and several Apocynoids colleters in a continuous ring, in multiple rows in some Tabernaemontaneae and Hunterieae, or a single antesepalous colleter (especially in Echiteae), and in several genera of Rauvolfioids and Apocynoids colleters are absent; corolla sympetalous, rarely apopetalous (a few Ceropegieae), salverform, infundibuliform, tubular, urceolate or rotate, lobes almost always 5 (very rarely 4), usually contorted in bud, either dextrorse or sinistrorse, more rarely valvate; corolline or gynostegial coronas often present; stamens 5 (rarely 4), filaments mostly straight, sometimes geniculate, sometimes connate around the style (some species of Forsteronia, Thoreauea), sometimes coiled around the style (Dewevrella, some species of Parsonsia and Thenardia), inserted on the corolla tube, on prominent staminal feet (broadened filament base fused with corolla tube) or forming a staminal tube, included to exserted; anthers introrse, rarely latrorse, in almost all Apocynoids, Secamonoideae and Asclepiadoideae with highly elaborated and lignified guide rails (lignified guide rails absent in most Rauvolfioids and in Periplocoideae) and often with an apical connective appendage, thecae 4, unequal in most Apocynoids, with dorsal ones smaller through presence of guide rails, reduced to 2 in Asclepiadoideae, dehiscence longitudinal, attached to the style-head forming a gynostegium (gynostegium absent in Rauvolfioids); nectaries in alternistaminal pockets on the staminal tube, on sides of staminal feet or 5 (rarely 2) lobes encircling the base of the ovary, these often fused to varying degrees into an (often deeply lobed) ring (in some Rauvolfioids and early-branching Apocynoids nectaries are adnate to the outer wall of the ovary at the base or are sometimes nonfunctional or absent); gynoecium normally of two carpels (very rarely up to five); ovary mostly apocarpous, sometimes congenitally (Rauvolfioids only) or postgenitally syncarpous (several Apocynoids), in some genera only one carpel developing, superior to subinferior; placentation marginal when the ovary is apocarpous, parietal or axile when syncarpous, when apocarpous upper part of the carpels fusing postgenitally to form a complex style-head that produces adhesive for pollen transport, with a pollen-trapping basal collar and/or pollen-presenting upper crest present in many Rauvolfioids and Apocynoids; stigma mostly on the underside of the style-head, often restricted to five chambers behind the guide rails, but style-head scarcely morphologically differentiated and nearly uniformly receptive in some Rauvolfioids; adhesive a sticky foam or mucilage, or differentiated into five translators with a scoop-like pollen receptacle and sticky base, or as five hard clips (corpuscles) usually accompanied by five pairs of flexible arms (caudicles) forming a pollinarium. Fruit in Rauvolfioids diverse: drupes, berries, follicles or capsules; seeds usually without a coma, naked, arillate, or winged or fimbriate at the margin very rarely with a coma (Haplophyton); in the remainder of the family, fruit almost always a pair of ventrally dehiscent follicles (often only one due to abortion or due to postgenital fusion; rarely a septicidally dehiscent capsule) with small seeds with a micropylar coma, rarely with a chalazal coma, coma at both ends (only in early-branching Apocynoids), or fringed with long trichomes circumferentially (a few Periplocoid and Hoya species), or without a coma.
Full-text available
The concerted evolution of morphological and behavioral specializations has compelling examples in ant castes. Unique to ants is a marked divergence between winged queens and wingless workers, but morphological specializations for behaviors on the ground have been overlooked. We analyzed thorax morphology of queens and workers in species from 21 of the 25 ant subfamilies. We uncovered unique skeletomuscular modifications in workers that presumably increase power and flexibility of head–thorax articulation, emphasizing that workers are not simply wingless versions of queens. We also identified two distinct types of queens and showed repeated evolutionary associations with strategies of colony foundation. Solitary founding queens that hunt have a more worker-like thorax. Our results reveal that ants invest in the relative size of thorax segments according to their tasks. Versatility of head movements allows for better manipulation of food and objects, which arguably contributed to the ants’ ecological and evolutionary success.
Full-text available
A diet consisting of agar, whole egg, honey, vitamins, and minerals was found to be satisfactory for rearing 28 species of ants representing 4 subfamilies of Formicidae.
Full-text available
All ants live in perennial colonies that exhibit three phases, namely foundation (initiation of new colonies), growth (production of more workers), and reproduction (production of sexuals). A greater understanding of colony life cycles is attained by contrasting the two main strategies of colony founding: independent and dependent colony foundation. During independent colony foundation, winged queens disperse by flight and initiate a colony alone, they feed the first brood of workers using energy provided by their metabolic reserves (claustral), or by foraging (non-claustral), and colonies must grow to a large size before being able to reproduce, which may take years. During dependent foundation, young queens are accompanied by a group of workers, which increases their chance of survival and allows for earlier production of sexuals, but also limits queen dispersal. The shift from independent to dependent foundation occurred many times in independent ant genera.
Full-text available
The overview of ant-plants in Peninsular Malaysia presented here covers: (i) diversity of participating ant and plant taxa; (ii) specificity of associations; (iii) morphological diversity of plant structures used for ant-housing; (iv) plant growth habit; and (v) food type of ants directly or indirectly derived from host plants. Included are unpublished observations both from field and herbarium studies and records of ant-plants obtained from a literature survey. It is the first comprehensive account of Malayan ant-plants on species level since many decades. At least 45 'true' myrmecophytic species belonging to 20 genera and 14 families are recognized. Twenty-two additional species (from 17 genera) are looked upon as potential myrmecophytes but evidence is still anecdotal and incomplete. Pasoh Forest Reserve is home to at least one third of the known myrmecophytes occurring on the peninsula. [*comment by Joachim Moog (23rd June, 2003): Unfortunately, some new 'print errors' occur in chapter 33, table 1, of the Pasoh book. Contrary to the proofs (which were ok) the table was reformatted again, resulting six times in uncomplete information in the "ant taxa" column (ant names are trimmed off). Also, the ant taxa specialized on living in certain host plants are not printed in bold (as they should, see caption to table 1), thus an important information has been unintentionally removed.]
For several decades, social insect research has been dominated by a "top-down" approach that begins with evolutionary theory or mathematical models. A "bottom-up" approach based on a detailed description of the physical, numerical and life history attributes of social insect colonies has been largely neglected. I define the quantitative description of colony attributes as sociometry, the measuring of a society. I argue that sociometry can be a generous, unbiased source of testable hypotheses, and leads to a deeper understanding of social insect function, life history and evolution. Whereas there is a large deficit in sociometric data, the deficit of colony ontogeny data, defined as sociogenesis, is even greater. Yet, social insects offer an opportunity to generalize developmental processes to the colony level. Moreover, these processes can be anchored in local ecological conditions, thus linking development to evolution. A simple, practical method for the simultaneous collection of sociometry/sociogenesis data is described. By complete sampling and measurement of the full size range of a focal species' colonies on several carefully chosen dates throughout the annual cycle, a description (sociometry) of colonies during growth (sociogenesis) and through the seasons (annual life cycle) is generated. Our understanding of social insect biology would be greatly enhanced by the widespread adoption of the sociometric/sociogenesis method as the starting point of social insect studies.
Hydnophytum formicarium, Myrmecodia tuberosa, Phymatodes sinuosa, and Dischidia rafflesiana are described as epiphytic myrmecophytes in low-productivity vegetation, growing on Sarawak white sand soils, which are fed by the relatively unaggressive ant Iridomyrmex myrmecodiae. The ants place large quantities of insect parts in the plants, which probably take up their decomposition products. Virtually all epiphytes in this community are trophically associated with a colony of I. myrmecodiae and one or more myrmecophytes, and it appears that there are not enough nutrients to support epiphytes unless they obtain extra nutrients from ants. Two conspicuous plant parasites, Dischidia gaudichaudii and Pachycentria tuberosa, take root in the debris dumps. In the light of these findings, it seems likely that the neotropical myrmecophytes with relatively unaggressive ant occupants are involved in a similar relationship. Experiments are needed to compare seed production or growth rate in epiphytic myrmecophytes that are occupied and unoccupied by I. myrmecodiae.
ALTHOUGH ant-plant mutualisms have been described in many ecosystems, the magnitude of the direct benefits from such relationships are hard to quantify. In Bako National Park, Sarawak, Malaysia, stunted 'kerangas' forests occur on nutrient-poor sandstone hills1-3. As trees are widely spaced and have a sparse leaf area, a significant amount of light reaches the tree trunks and enables a diverse community of epiphytes to thrive there4. One of these epiphytes, Dischidia major (Vahl) Merr. (Asclepiadaceae), has evolved unusual methods for enhancing carbon and nitrogen acquisition. We show here that a mutualistic relationship exists between ants of the genus Philidris and their host, D. major. Using stable isotope analysis, we calculate that 39% of the carbon in occupied host plant leaves is derived from ant-related respiration, and that 29% of the host nitrogen is derived from debris deposited into the leaf cavities by ants.
The colonial system and territoriality of Philidris (Formicidae; Dolichoderinae) ants occupying dominantly epiphytic myrmecophytes, Hydnophytum moseleyanum (Rubiaceae: Hydnophytinae), in a mangrove forest in Papua New Guinea were investigated using a test measuring ant aggressiveness. One ant colony occupied several H. moseleyanum plants on plural mangrove trees. The ants were polydomous and monogynous as the colonies always had a single queen. It was suggested that the arboreal ant fauna in the mangrove forest canopy revealed an ant mosaic distribution pattern. As the main food resource, the Philidris ants obtained honeydew secreted by the diaspidid scale insects on the shoot tips of the host mangrove trees of H. moseleyanum.