Journal of lnsect Behavior, Vol. 10, No. 3, 1997
Foraging Behavior on Carcasses in the Necrophagic
Fernando B. Noil I
Accepted January 24, 1997; revised March 6, 1997
necrophagy; foraging behavior; Meliponinae; Apidae; stingless bees.
Insects Collectively exploit a wide range of food sources, but the diets arc limited
to one or a few kinds of food in different lineages. They generally encompass
carbohydrates, proteins or amino acids, and salts (Roubik, 1989). In bees (Apoi-
dea), for example, floral products, nectar (carbohydrates), pollen (proteins), and
oils, arc their principal food components (reviewed by Wcislo and Cane, 1996).
However, they can also collect urine, feces, and animal carcasses searching for
mineral salts, water, and organic compounds (Baumgartner and Roubik, 1989).
In stingless bees (Meliponinae), some variations occur. Besides those cited
by Schwarz (1948) we have sugars from fruit pulp collection by Trigona spp.
(Baumgartner and Roubik, 1989), membracid honeydew by Trigona, Oxytri-
gona, Apis, etc. (Larcoa and Sakakibara, 1976; Cortopassi-Laurino, 1977; Cas-
tro, 1975; Mariconi, 1963), and extrafloral nectaries (Slansky and Rodriguez,
1987; Roubik, 1989; Roubik et al., 1995; Noll et al., 1996); proteins from
soybean bran by Geotrigona inusitata and fungi spores by Apis (Kerr, personal
observation); and salts from blood and carcasses by Apis (Chance, 1983; Crcwe,
1985). Knowledge about such habits is limited, and the cases listed above prob-
ably include some sporadic events, in which honey and pollen restriction involves
temporary exploitation of uncommon sources. Thus, the only group which dis-
plays fixed unusual feeding habits is the obligate necrophages [Trigona hypogea
group (Camargo and Roubik, 1991)]. They have completely replaced pollen by
animal flesh as a protein source (Roubik, 1982; Camargo and Roubik, 1991;
I Departamento de Biologia, Faculdade de Filosofia CiSncias e Letras de Ribeir~o Preto, Univer-
sidade de Silo Paulo, 14040-901 Ribeir~o Preto--SP, Brazil.
0892-7553/97/05004)463512.50/0 © 1997 Plenum Publishing Corporation
Noll et al., 1996). In addition, T. hypogea and T. necrophaga collect sugars
from fruits and extrafloral nectaries (Noll et al., 1996; Roubik et al., 1995). In
this aspect, such marked behavioral change as observed in the T. hypogea group
is an evolutionary novelty, implying ethological modifications which are not
well detailed. We attempted, in this work, to characterize the general foraging
process performed by T. hypogea and some details as the influence of size of
carcasses on the foraging process and source fidelity.
The observations were based on a Trigona hypogea colony transferred (8
Aug. 1989) from Cajuni (S,~o Paulo State, SE Brazil) to our laboratory at the
University Campus at Ribeirio Preto, 60 km away, where no other T. hypogea
colony was present. After collection, the nest was introduced into glass-covered
a wooden box, and the nest temperature was maintained at 28-30"C. A trans-
parent plastic tube connected the colony to the outdoors. If necessary, honey
was put in the opened storage pots and meat (pig kidney and liver) was put
outside the colony to collected by the workers in so that food always be stored
in the colony.
In order to standardize our observations, baits of pig kidneys were cut as
cubes with different specific sizes (volumes: 30, 50, 60, 100, and 120 cm 3) and
put outside the colony on glass plates. When dead animals (birds, toads, lizards,
rats, snakes) were found in the field and their tissues were well preserved, they
were collected and preserved at low temperatures (- 10°C) for use in the for-
aging behavior observations (see below).
All observations were performed directly during sunny days, with mean
temperatures around 28°C. Three types of experiments were done. (1) Foraging
behavior was analyzed during 13 days (17, 22, 24, 28, and 30 Sept. 1992; 18,
20, 22, 25, 27, and 30 Oct. 1991; 1 and 2 Nov. 1992) after distributing carcasses
(see Materials) 25 m away from the nest. (2) In order to estimate recruitment
of workers in sources of different sizes, kidney baits (volume--30, 60, or 120
cm3; see Materials) were put in the field isolated using three repetitions (7, 8,
9, 14, 15, 16, 21, 22, and 23 Nov. 1992). (3) Food fidelity was tested using
two kidney baits (50 and 100 cm3; see Materials) with two repetitions (4 and
11 Dec. 1992). First, the smaller was placed and, after the foraging process had
started, the larger was placed 10 m distant. In experiments 2 and 3, the baits
were observed during 9 (0900-1800).
Fora~ng Behavior in Tdgona hypogea
RESULTS AND DISCUSSION
Among the known methods of bee communication, the scent trail is one
of the most efficient to explore infrequent, rapidly deteriorating (it was observed
that the necrophagous bees avoid deteriorated, stinking meat), and very much
disputed food sources as the ones used by necrophagous bees that explore animal
carcasses for obtaining protein (Roubik, 1982). As reported in several Trigonini
(Kerr, 1969; Roubik, 1989) source location and exploration in necrophagous
stingless bees also involve tracing a spotted scent trail from the source to the
nest. After placing the carcasses for study (see Methods), in all 15 observed
cases, only one forager found the spot and started the scent trail. After finding
the spot the first scent marks were made on the proper carcass. Then the bee
landed on nearby leaves and twigs and departed for the nest. For marking,
workers bit the bait with their mandibles many times and flew close to leaves
(20 to 50 cm). On these leaves they rubbed their posterior legs and abdomen
and bit the leaves' edges. Similar marking behavior has been observed in
(Kerr and Rocha, 1988) and some social wasps (Jeanne, 1981).
Subsequently, the bees usually performed another flight [10 to 18 s, 13.6 + 4.4
s (mean + SD); n = 30], returning to the spot and restarting the process four
to eight times (6.2 + 1.3; n = 15). Later the incoming foragers followed the
marked route. It is interesting that, in some cases, the eusocial epiponine wasp
followed the trail laid by
because they were seen following the same spots that had been previously scent-
marked by the bees. However, once on the source, they usually explored parts
not visited by the bees, probably a form of avoiding aggression.
Under normal conditions, i.e., sunny days, from 19 to 59 min (38.3 +
14.9 min; n = 15) was enough to attract the first scout-bee, the mass of foragers
(20 to 41 workers, 27.9 + 7.5 workers; n = 15) arrived in between 22 and
120 min (51.3 4- 28.7 min; n = 15), and this number increased rapidly to 40
to 108 foragers (74.3 4- 25.7 foragers; n = 15). Rapidly the source area was
filled by workers. The proportion was about 1 to 1.94 workers/cm 2 (1.5 4- 0.3
workers/cm2; n = 15). In fact, by this recruitment method, even small sources
of food could be exploited by T.
In one instance, the brood of a fallen
wasp nest of
was being collected, 800 m away from colony.
Roubik (1982) described the methods involved in food collection by nec-
rophagous bees. In addition, we noticed the deposition of regurgitated material
encircling the spot of collection. It involved viscous drops produced after ingest-
ing and regurgitating parts of the substratum being explored. The meaning of
such depositions is unknown. But such drops seem to be a kind of territorial
mark aimed at protecting the source against competitors, because flies and ants
that contacted these marks displayed an avoidance reaction. The length of time
I. Variation in the Number of
Workers Observed on Pig Kidney Baits of
Bait Mean 5:
SD Diff. of
size (cm 3) (em 3) Comparison means t value P < 0.05
30 26.43 + 21.26 30 vs 120 em 3 -72.43 -7.32 Yes
60 67.29 + 33.73 30 vs 60
-40.86 -4.13 Yes
120 98.86 + 50.19 60 vs 120
-31.57 -3.19 Yes
a worker collected food from carcasses ranged from 20.0 to 46.0 min (31.5 +
6.0 min; n = 15). After returning to the nest, the material transported in the
crop went directly into the pots.
The number of workers recruited for foraging differed according to the
source size (Table I). Baits of different sizes were statistically different in relation
to the number of recruited bees. No variation, i.e., not statistically significant
variation, in forager number was detected when a larger bait was placed 10 m
away from the forage bait (Wilcoxon's test, Z = 4.41). This suggests that when
a group of foragers is recruited, they keep collecting the source even if a larger
one is found, and only other foragers can be recruited to the new source.
In spite of its unusual independence from flowers, the foraging methods
to collect its odd food are not different from the ones widely
used by other Meliponinae, especially because they share some similarities with
other stingless bees concerning the use of a scent trail between the nest and the
source (Kerr, 1969).
Basically the foraging process can be divided into three distinct sequential
phases: (a) recognition--in general, a single worker spots the food source; Co)
establishment--mass recruitment involves intense agitation among foragers flying
around the spot, landing on the scent trail, and chasing away other insects; and
(c) foraging--intense food collection uses as many workers as possible for pro-
viding fast gathering. In this period the worker number is maintained and the
The number of foragers on carcasses seems to be related to the source size.
In addition, when some baits were put out simultaneously, there was no apparent
interference in the previous recruitment because the foraging patterns were sim-
ilar to those observed when only one bait was placed.
Foraging Behavior in
This paper was aided by Grant 91/3046-0 from Fapesp (Fundaq~o de
Amparo h Pesquisa do Estado de S~o Paulo). The author especially thanks
Ronaldo Zucchi, William T. Wcislo, and two anonymous referees for their
reading of the manuscript and helpful suggestions. Special thanks go to Sidnei
Mateus for his unselfish help in the field.
Baumgartner, D. L., and Roubik, D. W. (1989). Ecology of necmphilous and filth-gathering sting-
less bees (Apidae: Meliponinae) of Peru.
J. Kans. Entomol. Soc.
Camargo, J. M. F., and Roubik, D. W. (1991). Systematics and bionomics of the apoid obligate
group (Hymenoptera: Apidae, Meliponinae).
J. Linn. Soc.
44: 13-39. .
Castro, P. R. C. (1975). Mutualism entre
(Fabricius, 1793) e
Spreng. na presenga de
Ci~ncia e Cultura
Chance, M. M. (1983). Honeybees observed feeding on the blood of a bear.
Cortopassi-Laurlno, M. (1977). Notas sobre associa96es de
Trigona (Oxytrigona) tataira
BoL ZooL Univ. S. Paulo
Crewe, R. M. (1985). Bees observed foraging on a impala carcass.
Bee World 66: 8.
Jeanne, R. (1981). Chemical communication during swarm emigration in the social wasp
Kerr, W. E. (1969). Some aspects for the evolution of social bees.
Kerr, W. E., and Rocha, R. (1988). Communicag~o em
Melipona rufiventris e Melipona compres-
sipes. Cidnica e Cultura
Laroca, S., and Sakakibara, A. M. (1976). Mutualismo entre
Trigona hyalinata branneri
Rev. Bras. Ent.
Mariconi, F. A. (1963).
Inseticidas e seu emprego no combate ds pragas,
Bibl. Agron6mica Ceres,
Nogueira Neto, P. (1970).
A criacao de abelhas indfgenas semferrao,
Editora Ch~icaras e Quintais.
Noll, F. B., Zucchi, R., and Jorge, J. A., and Mateus, S. (1996). Food collection and maturation
in the necrophagous stingless bee,
(Hymenoptera, Meliponinae). J.
Roubik, D. W. (1982). Obligate necrophagy in a social bee.
Roubik, D. W. (1989).
Ecology and Natural History of Tropical Bees,
Cambridge Unive~ity Press.
Roubik, D. W., Yanega, D., Aluja, S. M., Buchmann, S. L., Inoue, D. W. (1995). An optimal
nectar foraging by some tropical bees (Hymenoptera: Apidae).
Schwarz, H. F. (1948). Stingless bees (Meliponinae) of the western hemisphere.
Bull. Am. Mus.
Wcislo, W. T., and Cane, J. H. (1996). Floral resource utilization by solitary bees (Hymenoptera:
Apoideae) and exploitation of their stored foods by natural enemies.
Annu. Rev. Entomol.