Content uploaded by Hermógenes Fernández-Marín
Author content
All content in this area was uploaded by Hermógenes Fernández-Marín
Content may be subject to copyright.
RESEARCH ARTICLE
Biological notes on a fungus-growing ant, Trachymyrmex cf. zeteki
(Hymenoptera, Formicidae, Attini) attacked by a diverse
community of parasitoid wasps (Hymenoptera, Diapriidae)
B. Pe
´rez-Ortega •H. Ferna
´ndez-Marı
´n•
M. S. Loia
´cono •P. Galgani •W. T. Wcislo
Received: 10 March 2009 / Revised: 13 November 2009 / Accepted: 18 February 2010
!Springer Basel AG 2010
Abstract A number of wasps in the family Diapriidae,
subfamily Diapriinae (Proctotrupoidea), are parasitoids that
specializeon ant larvae. These wasps are abundantand diverse
in the Neotropics, but little is known about their biology. We
studied parasitism rates by an array of diapriine wasps that
attack the larvae of fungus-growing ants, Trachymyrmex cf.
zeteki, in a single population (near Gamboa, Panama
´). Rela-
tively little is known about the biology and natural history of
these ants, so we also present data on colony size and nest
architecture. We excavated 136 colonies in central Panama
´
from June to September 2006, and 20 nests from July 2009.
We reared six wasp morphotypes; two of them in the genus
Mimopriella Masner and Garcia, one Oxypria Kieffer, two
Szelenyiopria Fabritius and one Acanthopria Ashmead. The
mean intensity of larval parasitism per ant colony was 33.9%
(2006), and its prevalence across all ant populations was
27.2% (2006 and 2009). Parasitism rates were not positively
correlated with host colony size. A single case of super-par-
asitism was documented in which two Oxypria males were
reared from the same host larva.
Keywords Diapriinae !Fungus-growing ants !
Parasitoid wasps !Adaptive radiations !Trachymyrmex
Introduction
The New World fungus-growing ants (Hymenoptera, Formi-
cidae, Attini) are especially diverse in the tropics (Weber,
1972). As is true for most social insects (Schmid-Hempel,
1998), they accumulate significant stores of resources within
their nests, attracting a diverse array of predators, microbial
pathogens, parasites, and parasitoids (e.g. Feener and
Brown, 1993; Schultz et al., 1998; Loia
´cono et al., 2000; La
Polla et al., 2002; Masner and Garcı
´a, 2002; Mueller et al.,
2005; Powell and Clark, 2004; Ferna
´ndez-Marı
´n et al., 2006,
2009). Considerable attention has been given to the
microbes attacking attine ants, but their insect enemies are
less well studied, except for phorid flies (e.g. Feener and
Brown, 1993).
Diapriine wasps (Hymenoptera, Proctotrupoidea, Diaprii-
dae) are mainly Neotropical, and are frequently associated
with ants; little is known of their biology and there are
numerous undescribed species (Loia
´cono et al., 2000;
Masner and Garcı
´a, 2002; Ferna
´ndez-Marı
´n et al., 2006).
The subfamily contains an estimated *1,000 species in
Costa Rica alone (Hanson and Gauld, 2006), so the taxo-
nomy is difficult. Some diapriines apparently mimic their
hosts’ morphology (e.g. Loia
´cono et al., 2000; Masner and
Garcı
´a, 2002), which presumably aids them in avoiding
detection by host ants, but there are few behavioral data on
host-parasitoid interactions.
We studied aspects of the intensity and prevalence of these
little-known diapriine wasps that attack the larvae of a fun-
gus-growing ant, Trachymyrmex cf. zeteki (Attini); we also
provide notes on the ants’ nest architecture, and host-parasi-
toid behavioral interactions. We discovered a remarkably
diverse community of parasitoids within one host population
near Gamboa, Panama
´. High rates of parasitism should
effectively slow the demographic growth rate of the ants,
B. Pe
´rez-Ortega !H. Ferna
´ndez-Marı
´n(&)!P. Galgani !
W. T. Wcislo (&)
Smithsonian Tropical Research Institute, Balboa,
Apartado 0843-03092 Anco
´n, Republic of Panama
e-mail: hermogenes_f@hotmail.com
W. T. Wcislo
e-mail: WcisloW@si.edu
M. S. Loia
´cono
Museo de La Plata, Divisio
´n Entomologı
´a,
Universidad Nacional de La Plata, Buenos Aires, Argentina
Insect. Soc.
DOI 10.1007/s00040-010-0086-1 Insectes Sociaux
by removing recruits to the work force, with potentially
important consequences for understanding social biology.
Materials and methods
Nest collections
Established colonies of Trachymyrmex cf. zeteki were col-
lected near Pipeline Road in Parque Nacional Soberanı
´a,
near Gamboa in central Panama
´(9"0703800N, 79
o
4300900W).
Collections included eight nests from June to August, 2005,
which were used only to obtain taxonomic identifications;
we collected 136 nests from June to September 2006 and 20
from July 2009. Trachymyrmex cf. zeteki were easily found
on the banks of creeks where they were abundant, as well as
on nearby slopes. Data are given as mean ±standard devia-
tions. Voucher specimens of ants and wasps were deposited
in the Museo de La Plata, Argentina; the Museo de los In-
vertebrados de la Universidad de Panama
´, Panama
´, and the
Dry Reference Collection, Smithsonian Tropical Research
Institute.
Nest architecture, demography and nesting biology
Nests were excavated using a hand pick and a small knife.
During excavations, we recorded the following features of
nest architecture. The auricle surrounding the entrance was
measured as the longest vertical axis and the shortest
horizontal axis, and its height was measured from the base
to the highest part. We measured the diameter and length of
the tunnel that led to the first chamber, and the width and
height of each chamber. We also noted the positions of
tunnels that connected adjacent chambers; the shape and
location of any structures that supported the fungus gardens
away from the soil chambers; the location of eggs and
brood within the garden; and the location of the nest in the
field. We collected the resident ants and brood using an
aspirator. After excavating each nest the entire garden and
all resident ants and brood were transported to the labo-
ratory and maintained in a plastic box for 2 months using
standard methods (Weber, 1972), and all brood that were
present at the time of collection were repeatedly examined
for evidence of parasitism until ants or parasitoids emerged
as adults, or until the brood died, using methods given in
Ferna
´ndez-Marı
´n et al. (2006).
Intensity and prevalence of wasp parasitism
For each nest we counted the number of queen(s), workers,
eggs, larvae, and pupae using a stereomicroscope. The
numbers of immatures and workers may be underestimates
because some may have been lost during collections, if
some immatures were buried in soil or adults escaped
running. We recorded the number of older larvae parasit-
ized by wasps, which were easily recognized by the dark
coloration through the cuticle of an ant larva in which a
wasp was developing (Ferna
´ndez-Marı
´n et al., 2006). For
some nests, larvae from separate chambers were main-
tained in the laboratory in separate containers, in order to
determine whether larvae in chambers that were closer to
the entrance were more heavily parasitized. When a para-
sitized larva was identified, it was removed and maintained
in a separate Petri dish until the wasp emerged or the larva
died. Data on parasitism rates were pooled across morpho-
species, because of difficulties in identifying morphotypes
with a high degree of accuracy coupled with low survival
rates of untended brood. These morphotypes will be descri-
bed in a forthcoming paper with Dra. M. Loia
´cono (Museo
de La Plata, Argentina).
Behavioral interactions between ants and parasitoids
Five nests of T. cf. zeteki were observed continuously in the
field from 0800 to 1600 between 21 and 28 September
2006 (when nests contained parasitoids and hence adult
females were active), each for 8 h to observe attempts by
wasps to invade the nests. In the laboratory, behavioral
interactions were recorded with a color CCTV camera
attached to a stereomicroscope and illuminated with fiber
optics, using minimal light levels. We subsequently viewed
the recorded behaviors on a monitor. We observed six
parasitized nests for *30 min each, and recorded the
number of ant larvae within the field of view. Each
recorded video was analyzed using scan sampling every
3 min: behaviors scored included contact with the larva or
passing it; grooming the larva; and moving it.
Results
Nest architecture, demography and nesting biology
Nests were located along creeks and on slopes, below roots,
stones and dead branches, which presumably provide protec-
tion against water run-off. Mature colonies of Trachymyrmex
cf. zeteki were conspicuous due to an auricle at the entrance,
made of packed soil (Fig. 1d). The auricle measured
28.6 ±7 mm along the vertical axis, 21.3 ±5.3 mm along
the horizontal axis and 16.2 ±8.5 mm in height (n=103),
and surrounded a single entrance that opened into a small
antechamber, which in turn opened into a narrow tunnel that
was 3.9 ±0.5 mm in diameter and 12.1 ±9.2 mm long
(n=11). There was a positivecorrelation between the number
of workers and the number of chambers (Pearson’s r=0.6;
B. Pe
´rez-Ortega et al.
P\0.0001; n=125). Nests with one or two chambers were
the most common (45.6 and 36.8% of 125 nests, respectively),
but some nests had three (16%) or four (1.6%) chambers.
The chambers were connected through a tunnel that opened in
the floor and led to an opening in the ceiling of the next
chamber. The chambers were generally semi-spherical
(width =35.3 ±10.2 mm, height =30.4 ±6.0, n=11)
with rootlets entering the chambers, which suspended the
fungus gardens free from the soil in 50%of the cases; in 30% of
the chambers the gardens wereattached directly to the soil floor
and in the remaining chambers (20%) the gardens were sus-
pended by living rootlets but contacted the soil at one point on
the floor. In all cases the juveniles, larvae and pupae were
located in the periphery of the garden, while the eggs were
distributed inside it (n=136).
For 136 nests excavated, 83 nests had one queen, and ten
nests had between two and four de-alate females; in 43
nests the queens were not found. The number of workers
varied from 3 to 397 individuals per colony (!
x=78.3 ±
66.2) and worker number correlated positively with queen
number (Pearson’s r=0.30, P=0.003, n=93). Eggs
and other brood were found in all months of this study and
there was a positive correlation between colony size and
total brood number (Pearson’s r=0.55, P\0.0001,
n=136). The number of colonies with alate ants was
greatest in June 2006; most of these ants were males,
except one colony that had 1 female and 11 male alates. In
July and September only one colony had alate ants and all
were males (Table 1).
Intensity and prevalence of parasitism by wasps
There was no significant correlation between colony size
and the proportion of parasitized larvae (Pearson’s r=
-0.01; P=0.9, n=136). Of 136 nests, 33 (24.3%)
contained parasitized larvae. Within nests 3.1–100% of
Fig. 1 a–c. Diapriidae parasitoids of Trachymyrmex cf. zeteki (Formicidae): aSzelenyiopria sp., female, scale bar =0.6 mm; bMimopriella
sp., female, scale bar =0.6 mm; cAcanthopria sp., male, scale bar =0.7 mm; dnest entrance auricle of host ant, scale bar =10 mm
Biological notes on a fungus-growing ant
larvae were parasitized, and the overall mean intensity of
parasitism was 33.9%. The nests with 100% parasitism
rates had relatively few adult workers (N=4, 9, and 41).
Prevalence and intensity per month are given in Table 2.
Six wasp morphotypes were reared from Trachymyrmex
cf. zeteki larvae; two of them belong to the genus Mimo-
priella (Masner and Garcia), one to Oxypria (Kieffer), two to
Szelenyiopria (Fabricius), and one to Acanthopria (Ash-
mead). Although our information is limited, some nests were
parasitized by more than one parasitoid wasp morpho-
species. In one nest we reared wasps of Mimopriella morpho-
species 1 and Szelenyiopria morpho-species 2; another nest
had wasps of three morpho-species including, Mimopriella
sp. 1, Szelenyiopria sp. 1, and Acanthopria sp. 1. Based on
nine colonies from 2009, we found one nest with one
Acanthopria wasp; five nests with Mimopriella sp. 1; one
nest with Mimopriella sp. 2; three nests with Szelenyiopria
sp. 1, and one nest with Szelenyiopria sp. 2. We found one
case of super-parasitism in which two males of one Oxypria
morpho-species were reared from the same larva. Overall
the wasp sex ratio reared from ants was female-biased
(1:2.72, m:f, n=119 wasps, 2006 only).
The nest entrance architecture was not associated with
overall parasitism rates. All correlations between the size
of the auricle and parasite intensity were insignificant
(vertical axis: r=-0.2, P=0.324; horizontal axis:
r=-0.12, P=0.563; height: r=-0.23, P=0.259;
n=27). Of nests with multiple chambers, 18.2% had
parasitized larvae in all chambers; 27.3% had them only in
the outermost chamber; 12.1 and 3% had them only in the
inner second and third chambers, respectively; and in 9.1%
of nests we could not determine which chambers were
infested because the chambers collapsed during excava-
tions.
Behavioral interactions between ants and parasitoids
During 41.5 h of observations in the field we never
observed diapriine wasps close to the ants’ nests, despite
rearing wasps from nests collected at the same time and
despite observing wasps in the field. Observations in the
field suggest that ants do not discriminate between para-
sitized and unparasitized larvae, because when the first
chamber that contained brood was opened, workers moved
both parasitized and unparasitized larvae and pupae to the
next chamber. In the laboratory, however, worker ants
contacted larvae with parasitoids more frequently than
those without parasitoids (5.2 ±2.87 and 1.9 ±1.33 con-
tacts per larva, respectively), and moved them more
frequently as well (1.2 ±1.73 and 0.58 ±0.95, respec-
tively) (n=40; 3 min scans). Sample sizes were small and
the significance of the behavior is unclear because the ants
did not destroy the parasitized larvae.
Discussion
The diapriine wasps, including Mimopriella, Oxypria,Szele-
nyiopria, and Acanthopria in our study are koinobiont
parasitoids of the larvae of Trachymyrmex cf. zeteki ants.
This finding provides further evidence that diapriine wasps
are associated with fungus-growing ants (Attini) (Kistner,
1982; Loia
´cono et al., 2000; Masner and Garcı
´a, 2002;
Table 1 Demography of Trachymyrmex cf. zeteki from 136 nests excavated along Pipeline Road, Gamboa, Panama
´. Data are given as mean
±SD with ranges given below in parentheses
Date (2006) Colonies N N N N
Collected With alates Eggs Larvae Pupae Workers
June 46 8 6.57 ±5.89 14.13 ±13.32 12.38 ±21.65 82.87 ±79.28
(0–28) (0–52) (0–67) (4–397)
July 44 1 9.66 ±6.48 13.89 ±8.61 10.89 ±10.82 64.75 ±38.88
(0–36) (1–38) (0–51) (9–196)
August 19 0 10.47 ±6.7 16.58 ±12.18 11.89 ±11.73 103.0 ±80.67
(2–26) (3–56) (0–43) (13–376)
September 27 1 8.56 ±6.63 13.19 ±13.00 8.22 ±10.32 75.15 ±63.65
(0–22) (0–59) (0–43) (3–253)
Table 2 Intensity and prevalence of diapriine wasps that attack T. cf.
zeteki fungus-growing ants along pipeline road, Gamboa, Panama
´
Month (2006) Colonies Prevalence Intensity
Collected Parasitized (%) (%)
June 46 10 21.7 47.4 (0–100)
July 44 15 34.1 33.3 (0–71)
August 19 3 15.8 17.1 (0–24)
September 27 5 18.5 19.3 (0–31)
July (2009) 20 9 45.0 –
Total 156 42 26.9 33.9
B. Pe
´rez-Ortega et al.
Ferna
´ndez-Marı
´n et al., 2006). Between *15 and 34% of
ant colonies were parasitized by wasps, and in several cases
involving small colonies, 100% of larvae were parasitized,
suggesting these wasps inflict a heavy cost on the ants,
roughly comparable to parasitism rates by diapriine wasps
that attacked Cyphomyrmex ants (Ferna
´ndez-Marı
´n et al.,
2006). To date, however, there are no comprehensive studies
that examine colony growth rates as a function of the number
of worker ants per colony, so the impact of these wasps
on colony demography remains to be determined. We lack
information concerning competition among parasites that
attack Trachymyrmex.
Some features of nest architecture are frequently assu-
med to play a role in defending stored resources against
natural enemies. Our data indicate that the nest entrance
structure is not associated with parasitism rates, unlike the
situation in species of sweat bees (Halictinae) or Camp-
onotus ants (Formicidae), for example, which construct a
narrow collar at the nest entrance that decreases the like-
lihood that predators and parasites enter the nest (e.g.
Sakagami and Michener, 1962; Smith et al., 2003; Powell,
2008). The use of multiple brood chambers may render
larvae in the outermost chamber more susceptible to parasi-
toids than larvae in inner chambers, although interpreta-
tions of data are complicated by the fact that ants some-
times move larvae between chambers. Some cavity-nesting
stingless bees (Meliponini) construct an empty cavity as a
false nest, adjacent to the nest entrance, and then construct
a second cavity further in, which is thought to deter
predators (Wille and Michener, 1973; for other taxa, see
Tepedino et al., 1979; Munster-Swendsen, 2000; Ası
´s,
2007).
Our scant data on behavioral interactions are inconsistent
with respect to whether ants are able to detect parasitized
brood, as inferred from rates of contact and movement
involving parasitized and unparasitized larvae, but our
sample sizes are too small to warrant firm conclusions.
Ramos-Lacau et al. (2007) report that Cyphomyrmex ants do
not discriminate between brood infested with Acanthopria
parasitoids versus uninfested brood. In our study ants did
not destroy parasitized larvae, suggesting the latter are not
recognized, nor is there any evidence that parasitoids
actively manipulate host behavior to enhance the fitness of
parasitoids.
We reared six morpho-species from four genera of
wasps in a single host population, which is a more diverse
assemblage than observed for those that attack Cypho-
myrmex (Attini) (Ferna
´ndez-Marı
´n et al., 2006). We found
some nests with more than one morpho-species reared from
the same colony, so the limited comparative data suggest
super-parasitism may not be unique to Trachymyrmex
(Loia
´cono et al., 2000; Ferna
´ndez-Marı
´n et al., 2006; Ramos-
Lacau et al., 2007). Trachymyrmex cf. zeteki are found in
the same habitats in central Panama as is C. rimosus, and
both have high parasitism rates, but they do not share di-
apriine morpho-species (compare Results with Ferna
´ndez-
Marı
´n et al., 2006). Detailed studies are needed on host
selection behavior to understand what factors might pro-
mote the coexistence of so many parasitoids in a single host
population, and what factors preclude coexistence within
host colonies (see ‘‘Discussion’’ in Ferna
´ndez-Marı
´n et al.,
2006).
Combined with earlier work (Loia
´cono et al., 2000;
Masner and Garcı
´a, 2002; Ferna
´ndez-Marı
´n et al., 2006),
our study hints that these tiny wasps play a significant role
in the population biology of attine ants, and thus warrant
the attention of biologists studying the roles that natural
enemies play in shaping the social evolution of attine ants
(e.g. Ferna
´ndez-Marı
´n et al., 2009).
Acknowledgments We are grateful to Adam Smith, Simon Tierney
and two anonymous reviewers for help with comments on the man-
uscript; to the Autoridad Nacional del Ambiente de la Repu
´blica de
Panama
´for research, collecting, and export permits; and to the rest of
the staff of the Smithsonian Tropical Research Institute (STRI) for
logistical support. BPO received financial support from the Internship
Program of the Office of Academic Programs at STRI; HFM was
supported by a Post-doctoral Fellowship from the Smithsonian
Institution, and a Postdoctoral fellowship from the Secretarı
´a Nac-
ional de Ciencia, Tecnologı
´a e Innovacio
´n, Panama
´. We gratefully
acknowledge additional funding from the F. H. Levinson Fund, and
general research funds from STRI to WTW.
References
Ası
´s J.D. 2007. The significance of the vestibular cell in trap nesting
wasps (Hymenoptera: Crabronidae): Does its presence reduce
mortality? J. Insect Behav. 20: 289-305
Feener D.H. Jr. and Brown B.V. 1993. Oviposition behavior of an ant-
parasitizing fly, Neodohrniphora curvinervis (Diptera: Phoridae),
and defense behavior by its leaf-cutting ants host Atta cephalotes
(Hymenoptera: Formicidae). J. Insect Behav. 6: 675-688
Ferna
´ndez-Marı
´n H., Zimmerman J.K. and Wcislo W.T. 2006.
Acanthopria and Mimopriella parasitoid wasps (Diapriidae)
attack Cyphomyrmex fungus-growing ants (Formicidae, Attini).
Naturwissenschaften 93: 17-21
Ferna
´ndez-Marı
´n H., Zimmerman J.K., Nash D.R., Boomsma J.J. and
Wcislo W.T. 2009. Reduced biological control and enhanced
chemical pest management in the evolution of fungus-farming in
ants. Proc. R. Soc. London, Ser. B. 276:2263-2269
Hanson P.E. and Gauld I.D. 2006. Himeno
´ptera de la Regio
´n Neo-
tropical. The American Entomological Institute. Gainesville, FL.
994 pp
Kistner D.H. 1982. The social insects’ bestiary. In: Social Insects,
Vol. III (Hermann H.R., Ed), Academic Press, New York. pp 1-
244
La Polla J.S., Mueller U.G., Seid M. and Cover S.P. 2002. Predation
by the army ant Neivamyrmex rugulosus on the fungus-growing
ant Trachymyrmex arizonensis.Insect. Soc. 49: 251-256
Loia
´cono M.S., Margaria C.B., Quiran E.M. and Corro Molas B.M.
2000. Dia
´pridos (Hymenoptera) parasitoides de larvas de la
hormiga cortadora Acromyrmex lobicornis (Hymenoptera: For-
micidae) en la Argentina. Rev. Soc. Entomol. Argent.59: 7-15
Biological notes on a fungus-growing ant
Masner L. and Garcı
´a J.L. 2002. The genera of Diapriinae (Hymeno-
ptera: Diapriidae) in the new world. Bull. Am. Mus. Nat. Hist.
268: 1-138
Mueller U.G., Gerardo N.M., Aanen D.K., Six D.L. and Schultz T.R.
2005. The evolution of agriculture in insects. Ann. Rev. Ecol.
Evol. Syst. 36: 563-595
Munster-Swendsen M. 2000. Interaction between the solitary bee
Chelostoma florisomne and its nest parasite Sapyga clavicornis-
empty cells reduce the impact of parasites. Ecol. Entomol. 25:
63-70
Powell S. 2008. Ecological specialization and the evolution of a
specialized caste in Cephalotes ants. Funct.Ecol.22: 902-911
Powell S. and Clark E. 2004. Combat between large derived societies:
a subterranean army ant established as a predator of mature leaf-
cutting ant colonies. Insect. Soc. 51: 342-351
Ramos-Lacau L.S., Delabie J.H.C., Bueno O.C., Villemant C., Prates-
Luz H., Andrade-Miranda M.M., Pereira de Oliveira G., Rodri-
gues Silva Jr M.R., and Lacau S. 2007. Estrate
´gia comporta-
mental de Acanthopria Ashmead (Hymenoptera: Diapriidae),
parasito
´ide de Cyphomyrmex transversus Emery (Hymenoptera:
Formicidae). Biolo
´gico (Sa
˜o Paulo), 69: 451-454
Sakagami S.F. and Michener C.D. 1962. The Nest Architecture of the
Sweat Bees(Halictinae): A Comparative Studyof Behavior. Univer-
sity of Kansas Press, Lawrence. 135 pp
Schmid-Hempel P. 1998. Parasites in Social Insects. Princeton
University Press, Princeton. 392 pp
Schultz T.R., Bekkevold D. and Boomsma J.J. 1998. Acromyrmex
insinuator new species: an incipient social parasite of fungus
growing ants. Insect. Soc.45: 457-471
Smith A.R., Wcislo W.T. and O’Donnell S. 2003. Assured fitness
returns favor sociality in a mass-provisioning sweat bee, Mega-
lopta genalis (Hymenoptera: Halictidae). Behav. Ecol. Sociobiol.
54: 14-21
Tepedino V.J., McDonald L.L. and Rothwell R. 1979. Defense
against parasitization in mud-nesting Hymenoptera: can empty
cells increase net reproductive output? Behav. Ecol. Sociobiol.6:
99-104
Weber N.A. 1972. Gardening Ants: The Attines. The American Philo-
sophical Society, Philadelphia. 146 pp
Wille A. and Michener C.D. 1973. The nest architecture of stingless
bees with special reference to those of Costa Rica (Hymeno-
ptera: Apidae). Rev. Biol. Trop. 21: 1-278
B. Pe
´rez-Ortega et al.
