Original article
Sounds inMelipona costaricensis (Apidae: Meliponini):
effect of sugar concentration and nectar source distance
Ingrid AGUILARa*, Daniel BRICEÑOb
a
Centro de Investigaciones Apícolas Tropicales, Universidad Nacional, PO Box, 475-3000 Heredia,
Costa Rica
b
Escuela de Biología, Universidad de Costa Rica, San José, Costa Rica
(Received 7 July 2001; revised 2 January 2002; accepted 13 February 2002)
Abstract – The principal finding presented here is that Melipona costaricensis produces sounds that vary sig-
nificantly in relation to varying sugar concentrations. Several components of the sounds produced, during the
bees’stay inside of the hive, were positively correlated with sugar concentration: song duration, probability of
sound production, duration of the long pulses and the relationship between short / long pulsed sound. Number
of pulses per song was negatively correlated with sugar concentrations. The duration of all sound pulses was
significantly correlated with the distance of the food source. Duration of the short pulses and interpulses were
significantly affected by the distance of the food source. Sounds may contribute to the recruitment of
nestmates to increasingly rich food supplies during the day. It is clear that aspects of the sound signals are cor-
related with profitability of food sources.
sound production / sugar concentration / distance / food source /stingless bees
1. INTRODUCTION
When a foraging social bee (Apidae:
Meliponini) finds food, it returns to the hive
and may recruit nestmates to that source.
Through recruitment, which implies some
mechanism of communication, newly avail-
able food sources can be exploited more effi-
ciently (von Frisch, 1967). The known
mechanisms for recruitment vary considerably
between species of Apidae. They include olfac-
tory, taste and acoustic cues (Lindauer and
Kerr, 1958; Esch et al., 1965). It has been found
that honey bees give information about direc-
tion and distance of food sources by perform-
ing dances on the vertical combs in the hive
(von Frisch, 1967). Esch (1964) and Esch et al.
(1965) demonstrated that sound signals are
also an important element of honey bee dances
and apparently they are an adaptation to danc-
ing at low light intensities inside the nest
(Kirchner et al., 1996). These signals are air-
borne sounds of high intensity produced by
dorsoventral vibrat ions of the wings
(Michelsen et al., 1987; Kirchner et al., 1988).
Their sounds consist of short pulses at a repeti-
tion rate of 15 Hz (Kirchner, 1993), they do not
have a long continuous pulse sound.
The duration of sound pulses during each
performance of both the round dance and the
waggle dance increases with distance between
the hive and the feeder (Esch et al., 1965;
Waddington and Kirchner, 1992; Kirchner and
Dreller, 1993). Sounds produced during round
and waggle dances are necessary for successful
communication (Michelsen et al., 1989;
375
* Correspondence and reprints
E-mail: iaguilar@una.ac.cr
Apidologie 33 (2002) 375–388
© INRA/DIB-AGIB/EDP Sciences, 2002
DOI: 10.1051/apido:2002023
Kirchner and Sommer, 1992). It has been sug-
gested that both dances additionally may con-
tain useful information about profitability of the
food source (Spangler, 1991; Waddington and
Kirchner, 1992; Kirchner and Dreller, 1993) and
sound signals emitted in the round dances of
Apis dorsata may contain useful information
about direction (Michelsen et al., 1992;
Waddington and Kirchner, 1992; Kirchner,
1993). In A. mellifera the duration of sounds was
positively correlated with the sucrose concen-
tration (Waddington and Kirchner, 1992) and
the probability of sound production was corre-
lated with the profitability of the food source in
A. dorsata and A. mellifera (Waddington and
Kirchner, 1992; Kirchner and Dreller, 1993).
The communication systems for recruit-
ment adopted in the eusocial stingless bees, dif-
fer from those of honey bees (Lindauer and
Kerr, 1958; Kerr, 1959). Stingless bees use sev-
eral cues to locate food sources: scent marks
along the route to the food source, odour of
nestmates and of the food source (Kerr, 1959).
They may also be guided to the source by an in-
dividual worker bee. Scent trails allow them to
indicate not only distance and direction but also
height, whereas honey bees only communicate
distance and direction (Wilson, 1971; Nieh and
Roubik, 1995).
Not much information is available about the
intranidal behaviour of foragers and recruits of
Melipona or Trigona species. Observations in
some species showed that foragers, once they
reach a good food source outside the nest, made
irregular excited movements inside the hive as-
sociated sometimes with sound production
(Kerr, 1959; Lindauer and Kerr, 1958; Esch
et al., 1965). Returning foragers of M. scutellaris
(Lindauer and Kerr, 1960), M. quadrifasciata,
M. merillae (Esch et al., 1965) as well as M.
panamica (Nieh and Roubik, 1995) produce
sound pulses. A distance related signal is pro-
duced in the hive by forager bees of M.
quadrifasciata (Esch et al., 1965; Esch, 1967),
M. merillae (Esch, 1967) and M. panamica
(Nieh and Roubik, 1998). Nieh and Roubik
(1998) found that sound signals were corre-
lated likewise with height. Although communi-
cation of distance by sound signals seems to be
clear (Nieh, 1998b; Nieh and Roubik, 1998),
Hrncir et al. (2000) did not find a significant
correlation with distance to the food source for
either M. quadrifasciata or M. scutellaris.
It is unclear if quality of the food source
(sugar concentration) is communicated by
sound signals in stingless bees. There are indi-
cations that M. panamica behaviour varied de-
pending on the sugar concentration: foragers
appeared to produce lower amplitude sounds at
1.0M sugar concentration, although this was
not quantified, and they turned 36% more
slowly than foragers at a 2.5M (Nieh, 1998b).
As the concentration of the sucrose solution de-
creased, the rate of recruitment decreased.
Nieh (1998b) also noted that at artificial food
sources, more foragers produced sounds as the
concentration increased. Jarau et al. (2000) ob-
served in M. quadrifasciata and M. scutellaris
that foragers collecting food from a source
0.75M did not show any kind of specific behav-
iour nor did they produce any kind of sounds
when entering the hive. These observations
suggest that there may be variation among
stingless bee species in the use of sound for the
recruitment of foragers.
The objective of this paper is to provide in-
formation on whether sound “signals” (pulse
duration, interpulse duration, song duration,
long and short pulses, number of pulses per
song and probability of singing), produced by
trained M. costaricensis foragers inside the
hive change in relation with different sugar
concentration and with different distances to
the food sources.
2. MATERIALS AND METHODS
2.1. Study site and colony
The study was carried out in Costa Rica during
January, August 1998 and April 1999 at the Centro
de Investigaciones Apícolas Tropicales (Heredia
province) and Universidad de Costa Rica (San José
province). Data were obtained from a colony of
Melipona costaricensis of about 600 individuals
collected in Horquetas de Sarapiquí, Heredia prov-
ince. It was installed in a wooden observation box
(70 × 20 × 20 cm) and brought to our building in
1998.
376 I. Aguilar, D. Bruceño
2.2. Effect of the sugar concentration
on sound production
Foragers of Melipona costaricensis (N = 5) were
trained to an artificial food source containing 2.0M
sucrose solution, which was placed 10 m from the
hive according to the experimental set up used by
von Frisch (1967). The bees were marked individu-
ally with paint on the thorax. The feeder station with
the sucrose solution consisted of a removable paper
yellow plate on a 65 cm iron stake and a drinking cup
with 6 holes glued on the plate. The concentration of
sucrose solution was changed between series, vary-
ing the order of the concentration from 2.0M, 1.0M
and 0.5M (total: 17 series), with about 15 min obser-
vation per concentration. We offered the sugar solu-
tion ad libitum on the top of the feeder into the
plastic cup. We refilled the cup at the feeder table
with fresh sugar solution before each observation
with the corresponding sugar concentration, after
having removed residue of the former solution. Re-
turning Melipona costaricensis foragers produced
sounds inside the hive. Sounds and behaviour of
bees were video recorded each time the trained bee
returned from feeding station.
Songs, defined as multiple long and short sound
pulses produced while the forager was inside the
hive, began when the singing forager entered the
hive, and ended only when it has finished dispensing
its nectar load to nestmates. A returning forager per-
formed only one song, and some foragers may per-
formed none. Here, the “rate” of song production
was the number of songs performed by an individual
nectar forager per observation period.
To analyze the probability of sound production,
we considered only the returning foraging flights
within the first six minutes, after the bee started to
visit the feeder table. The probability of song pro-
duction was calculated by dividing the number of
visits during which the returning foraging bee
(N = 4) produced sounds inside the hive by the total
number of returning foraging flights.
The behaviour of the returning bees was recorded
inside a clear plexiglass chamber (11 × 10 × 4 cm)
located between the hive and the entrance exit tube,
using a Video Camera (CCD V-801) equipped with
+6 close up lens so that a single bee filled the screen.
The sound was recorded through a Senheiser MZK
80 ZV microphone inserted into the Plexiglas cham-
ber from above (3.5 cm).
Recordings of sounds were imported from video
recordings into a Pentium II 350 MHZ computer using
a 32 bit card. The mean duration of pulses and interval
between pulses were measured using the real time
display in the program Avisoft
®
. The standard devi-
ation of these measurements was determined by
measuring the duration of 10 pulses and intervals in
each of 10 different songs. The average differences
were 2.0 ms in pulse duration and 2.2 ms in interval
duration. N refers to number of bees; n to the number
of songs and n to the number of sound pulses ana-
lyzed.
2.3. Effect of the distance of the food
source on sound production
Bee foragers of M. costaricensis were trained to
an artificial food source with a 2.0M sugar solution
(von Frisch, 1967). The bees were marked individu-
ally by different spots of paint and trained to collect
the food offered at a feeder placed at distances of 1 m
(N = 5), 10 m (N = 3), 30 m (N = 5), 50 m (N = 5),
55 m (N = 5), 70 m (N = 5), 90 m (N = 5), 100 m
(N = 10), 130 m (N = 3) and 150 m (N = 3) from the
hive. The training procedure, bee behaviour and
sound analysis were conducted as described above.
2.4. Statistical analysis
The statistics are based on the average of all
pulses in a song, obtained for a specific parameter
(sound duration, interpulse duration, song length, re-
lation between long and short pulses and probability
of pulsed sounds) for each respective concentration
and distance. We estimated the relationship between
the mentioned parameters and both the concentra-
tion and the distance to the food source by express-
ing one in terms of a linear function of the other
through a regression analysis. Significant differ-
ences among the three different concentrations were
analyzed for long pulses with a non-parametric
Mann-Whitney U-test. We used the Friedman Two-
Way analysis of variance to test if the repeated mea-
sures on the same bees were significantly different in
probability of sound production. Likewise, we used
the coefficient of determination as an estimate of the
proportion of the variation of probability of sound
production determined by the variation of the sugar
concentration. We selected an alpha level of
P < 0.05.
3. RESULTS
3.1. General patterns of sound
production by foragers
The carrier frequency and temporal struc-
ture of a typical song of returning bees is shown
in Figure 1 (A and B). Returning nectar forag-
ers of M. costaricensis perform a song which
consists of multiple long and short pulses of
Sounds in Melipona costaricensis 377
relatively loud sound (Fig. 1). The song of M.
costaricensis has acoustical properties similar
to songs produced by other stingless bees (e.g.
M. quadrifasciata, M. scutellaris and M.
panamica). The carrier frequency of the signals
in M. costaricensis (Fig. 1A) ranged from 429
up to 588 Hz (493 ± 50 Hz, n = 30). We also
found up to eight peaks in the spectrum gener-
ated by other bees humming inside the hive. We
did not attempt to analyze in detail the fre-
quency of songs.
M. costaricensis did show a sequence from
long pulsed to short pulsed sounds (we will re-
fer to pulses shorter than 0.1 s as “short pulses”
and to pulses longer than 0.1 s as “long pulses”)
(Fig. 1B). During each performance (one re-
turn of the trained forager bee inside the hive) a
long pulsed sound was made first. The trained
forager bee continued this long pulsed sound
until she found a nestmate or nestmates to give
her load. Then she produced long and short
pulsed sounds while unloading food
(trophallaxis). Soundless periods were noted
during trophallaxis, and more frequently while
the trained bee returned to the feeder place.
Trained bees did not always produce sounds at
a 0.5M sugar solution. Returning pollen gath-
ers did not produce sounds at any time during
the period they were recorded.
Recruitment was observed during our ex-
periments. A total of 10 new bees landed at the
feeder table during the sugar concentration ex-
periments. From these new bees, six arrived at
the feeder when it contained 2.0M sugar solu-
tion. We also observed recruitment at the feeder
table during distance experiments (1 m,
N = 65; 10 m, N = 5; 30 m, N = 13; 50 m,
N = 5; 70 m, N = 1 new bees). We did not ob-
serve the arrival of new bees at 100 m and
150 m.
3.2. Effect of the sugar concentration
We observed variations in the songs of bees
trained at different concentrations (Fig. 2) in a
feeder 10 metres North from the hive. The song
duration of 2992 sound pulses (long and short
combined) was analyzed from bees visiting a
food source of 0.5M (number of songs, n = 41),
963 pulses at 1.0M (n = 60) and 4566 pulses at
2.0M (n = 38). The duration of all sound pulses
was correlated with sugar concentration
(P = 0.0000001, r = 0.232) (Fig. 3). In addi-
tion, the duration of the long pulses was posi-
tively correlated with sugar concentration
(P < 0.0001, r = 0.150). At 0.5M, long pulses
averaged 0.164 ± 0.1 s (n = 134), at 1.0M they
averaged 0.262 ± 0.225 s (n = 219), and at 2.0M
they averaged 0.343 ± 0.381 s (n = 1230). Also,
significant differences between the long pulses
among different food concentrations were found
(1.0M vs. 2.0M, P < 0.0016; 0.5M vs. 1.0M,
P < 0.00001; 0.5M vs. 2.0M, P < 0.000001,
Mann-Whitney two-tailed test).
The percentage of long pulses increased
with the increase in sugar concentration (4.4%,
22.7% and 26.9% respectively). Likewise, the
proportion of short to long pulses increased
considerably with the increase on sugar con-
centration (0.0468, 0.280 and 0.368 at 0.5M,
1.0M and 2.0M respectively) (P < 0.000001,
r = 0.61) (Fig. 4). The number of pulses per
song was negatively correlated with the sugar
concentration of the food source (P < 0.000001,
r = – 0.69).
A positive relation between the duration of a
complete song (one visit of a trained bee inside
the hive) and sugar concentration of the food
source was also found (P < 0.0000001,
r = 0.69). A negative correlation between total
interpulse duration and sugar concentration
also was found (P = 0.000001, r = – 0.13)
(Fig. 5).
The probability of sound production in-
creased with the increase of sugar concentra-
tion in each trained bee (N = 4). Different sugar
concentrations had an effect on the observed
behaviour. There was a significant difference in
probability of sound production at different
sugar concentrations (df = 3, x
r
2
= 29.34,
α = 0.001). Sugar concentration may explain
97% (r
2
= 0.970, n = 125 songs, pooled data)
of the variation in probability of song produc-
tion.
During some trials at low sugar concentra-
tions, some bees did not visit the feeder table at
all, but stayed in the nest receiving food from
other foragers. Other bees stayed outside mak-
ing a few visits to the feeder until food was
again substituted with sugar solution of either
1.0M or 2.0M.
378 I. Aguilar, D. Bruceño
Sounds in Melipona costaricensis 379
F
ig
u
r
e
1
A
380 I. Aguilar, D. Bruceño
Figure 1. A song consisting of pulsed sound produced inside the nest by the stingless bee Melipona
costaricensis. A, frequency spectrum ranges from 429–588 Hz. B, the temporal structure of the song [ampli-
tude (mV) vs. time (s)]. The extended part of above signal is showing single pulses.
Figure 1B
3.3. Effect of the distance of the food
source
We analyzed pulses when the food distances
were 1 m (n = 37, n = 1229), 10 m (n = 30,
n = 2046), 30 m (n = 19, n = 1882),50 m
(n = 35, n = 1371), 55 m (n = 5, n = 291), 70 m
(n = 29, n = 2557), 90 m (n = 21, n = 924),
100 m (n = 26, n = 1017), 130 m (n = 62,
n = 2182) and 150 m (n = 32, n = 2099) from
the hive. The duration of all sound pulses was
significantly correlated with the distance of the
food source (P = 0.026, r = 0.18) (Fig. 6). We
analyzed long pulses for food distances 1 m
(n = 463, 0.40 ± 0.7 s), 10 m (n = 785,
0.32 ± 0.32 s), 30 m (n = 398, 0.32 ± 0.58 s),
50 m (n = 449, 0.58 ± 0.68 s), 55 m (n = 91,
0.22 ± 0.17 s), 70 m (n = 1035, 0.34 ± 0.42 s),
90 m (n = 395, 0.75 ± 1.35 s), 100 m (n = 346,
0.30 ± 0.38 s), 130 m (n = 760, 0.34 ± 0.57 s)
and 150 m (n = 726, 0.33 ± 0.43 s); and short
pulses for food distances 1 m (n = 766,
0.038 ± 0.025 s), 10 m ( n = 1261, 0.038 ±
0.25 s), 30 m (n = 1484, 0.034 ± 0.025 s), 50 m
(n = 922, 0.044 ± 0.024 s), 55m (n = 200,
0.034 ± 0.026 s), 70 m ( n = 1522, 0.040 ±
0.027 s), 90 m (n = 529, 0.040 ± 0.025 s), 100 m
(n = 671, 0.05 ± 0.02 s), 130 m (n = 1422,
0.05 ± 0.02 s) and 150 m (n = 1373, 0.04 ±
0.02 s).
Sounds in Melipona costaricensis 381
Figure 2. Examples of variation of spectogram (frequency vs. time) of the waveforms at different sugar
concentrations of the stingless bee Melipona costaricensis: A, 0.5M; B, 1.0M and C, 2.0M.
Duration of long pulses was not correlated
with distance of the food source (P > 0.01,
r = 0.004) while the duration of the short pulses
showed a significant relationship (P < 0.01,
r = 0.094). The relation between short and long
pulses did not increase with the increasing dis-
tance from hive (0.60, 0.62, 0.26, 0.48, 0.45,
0.68, 0.74, 0.51, 0.53, 0.52 respectively)
(P > 0.0539, r = 0.114). Moreover, the per-
centage of long pulses neither increased with
longer distances (37.67%, 38.37%, 21.15%,
32.75%, 31.27%, 40.48%, 42.75%, 34.02%,
34.83%, 34.59% respectively) nor with the per-
centage of short pulses (62.33%, 61.63%,
78.85%, 67.25%, 68.73%, 59.52%, 57.25%,
65.98%, 65.17% and 65.41% respectively).
We found a positive correlation between the
duration of all interpulses and the food source
distance (P = 0.0014, r = 0.025) (Fig. 7). On
the other hand, the duration of a complete song
(one performance inside the hive) was neither
correlated with the food source distance
(P < 0.086, r = – 0.1) nor the number of pulses
per song (P > 0.3172, r = – 0.06). Number of
pulses dropped as the distance to the feeding
station increased (P < 0.00238, r = – 0.18).
Thus, distance did significantly affect some of
the acoustical variables.
4. DISCUSSION
It has been shown that acoustical signals are
used among stingless bees for communication
of distance and height of the food sources (Esch
et al., 1965; Nieh and Roubik, 1998). However,
different aspects of sounds are not correlated
consistently with distance among the stingless
bee species (Jarau et al., 2000).
As in M. scutellaris, M. quadrifasciata
(Jarau et al., 2000) and M. panamica (Nieh,
1999), foragers of M. costaricensis produced
long and short pulsed sounds. In M. seminigra,
M. quadrifasciata (Esch et al., 1965) and M.
panamica (Nieh and Roubik, 1998) foragers
produced pulsed sounds, whose durations were
positively correlated with the distance to the
food source. However, M. costaricensis did not
show the same behaviour. Nieh and Roubik
(1998) made a more detailed analysis in
382 I. Aguilar, D. Bruceño
Figure 3. The relationship between the sound duration (total of all sound pulses) and the sugar concentration
of the food source.
Sounds in Melipona costaricensis 383
Figure 4. Relation between pulsed sounds < 0.1 s and pulsed sounds > 0.1 s with increasing sugar concentra-
tion of the food source.
0.2 0.6 1.0 1.4 1.8 2.2
Figure 5. The relationship between the duration of all interpulses and the sugar concentration of the food
source.
M. panamica and concluded that sound signals
emitted during unloading of food give informa-
tion about height, and that sounds emitted dur-
ing a “dance” give information about distance.
Jarau et al. (2000), by contrast did not observe
in M. quadrifasciata and M. scutellaris the
dance behaviour mentioned by Nieh (1998a) in
M. panamica. Instead, they discovered that
another behaviour of returning bees, called
“jostling”, was correlated with recruitment
behaviour.
In the present study we found that trained
bees of Melipona costaricensis produced
sounds that varied significantly in relation to
the variation of sugar concentration in the food
source. When the quality of the food source
was high, as in the experimental feeder filled
with a 2.0M sucrose solution, all trained bees
continued visiting the feeder, worked quickly,
and produced strong sounds. When the quality
was low (0.5M) most of the bees stopped visit-
ing the feeder, and the ones that continued visit-
ing the feeder foraged slowly, spent more time
inside the hive between trips to the feeder, and
performed fewer pulses. Such modifications of
foraging behaviour in relation to the food
source profitability have been characterized in
honey bees (von Frisch, 1967; Núñez, 1966,
1970, 1982; Seeley, 1995) and recently have
been observed in stingless bees (Jarau et al.,
2000). Our findings confirm that workers of
Melipona (Biesmeijer et al., 1999) are able to
distinguish sugar solutions of different concen-
trations.
Spectrograms of the waveforms showed that
returning foragers produced pulsed sounds that
were very loud when they visited the feeder
with the highest concentration (Fig. 2). The
variation in this behaviour probably was
caused by the higher frequency levels that the
sound signals reached at greater sugar concen-
trations. The number of long pulses also in-
creased with sugar concentration. Sugar
concentration influences the body temperature
of honey bees, and consequently their wingbeat
frequency (Spangler, 1991; Waddington and
Kirchner, 1992). In stingless bees more ex-
periments are needed to confirm if sound
production is related to changes in body tem-
perature.
384 I. Aguilar, D. Bruceño
Figure 6. The relationship between the duration of all sound pulses and the distance between the colony and
the food source.
Our study shows that the variation in the
probability of producing a song in M.
costaricensis was explained by sugar concen-
tration of the food source (97%, r
2
= 0.970). In
A. mellifera and A. dorsata (Waddington and
Kirchner, 1992; Kirchner and Dreller, 1993)
this correlation suggests an alternative way by
which profitable food sources are signaled.
According to our data, other aspects of
sound signals, such as song duration and pro-
portion of long pulses, are potential signals of
richness of the food source in terms of sugar
concentration for M. costaricensis and likely
for others Melipona species. Firstly, this is sup-
ported by the finding (Nieh, 1998b) that
nestmates of M. panamica preferably contact
the abdominal region of a forager when a
source offers food with sugar in high concen-
tration. Hrncir et al. (2000) also found that re-
cruits of Melipona scutellaris and M.
quadrifasciata preferably contact a successful
forager at the thoracic-abdominal region. That
region is mentioned by Michelsen et al. (1987)
as a region where air particle movements are
more intensive in the waggle dance of
A. mellifera. Secondly, Nieh (1998a) points out
another interesting behaviour in M. panamica:
at the feeder table, significantly more foragers
produced sounds as the concentration of sucrose
increased (from 0.50M to 2.50M), and the pro-
portion of departures preceded by sounds in-
creased linearly. Although Nieh did not record
data on sounds inside the hive related with
sugar concentration, he found that recruitment
was higher at the feeder with the higher concen-
tration of sugar. Increasing recruitment at
higher concentrations also was found in
Trigona fulviventris (Johnson, 1987). Thirdly,
it appears that Melipona species prefer high
concentrations (Roubik et al., 1995; Biesmeijer
et al., 1999) and nectar of high sugar concentra-
tion becomes gradually more available through
the day (Roubik, 1984). This finding parallels
studies of daily patterns of nectar foraging, in
Melipona fasciata (Biesmeijer, 1997).
In honey bees, a bee reports the profitability
of a food source by adjusting the strength of her
dance. This adjustment could involve modula-
tion of signal duration (number of waggle runs
per dance) or signal intensity (sounds) or both.
Sounds in Melipona costaricensis 385
Figure 7. The relationship between the duration of all interpulses and the distance between the colony and
the food source.
A colony of honey bees achieves high foraging
success by allocating foragers simply in pro-
portion to the amount of dancing for each
source (Seeley, 1986, 1995). Why do stingless
bees, using just sound signals, not make use of
a similar method? Stingless bees do not have
dances comparable with those of honey bees,
but they do communicate direction or distance
mainly outside of the hive by laying phero-
mones, a scent beacon, using the food odor, or
by direct piloting. Inside the hive, it could be
more important how fast new bees can be moti-
vated and recruited to the available food
source.
Stingless bee foragers returning to the hive
from flower patches bring home information
about multiple food sources. Only bees return-
ing from highly profitable sources (e.g.
1.0M–2.0M) perform sounds. If we assume
sound is a recruitment signal, foragers will in-
tensify the rate of recruitment as the probability
of song production increases with the availabil-
ity of rich food sources, and as more foragers
discover the rich food source more sounds sig-
nals will be emitted. The singers will share
their information to bees standing near the en-
trance. Acoustical signals allow the nestmates of
the stingless bees to perceive and distinguish the
richer and poorer nectar collectors through the
darkness of the hive (Kirchner, 1993). Recruit-
ment, to pollen sources probably uses other
strategies because pollen collectors of M.
costaricensis and M. favosa never produced
pulsed sounds inside the hive during our obser-
vations, but instead went straight to the pollen
pots.
The relationships between decreasing
sound pulses with distance to the food source
confirm the relationship between the sound
signals with recruitment to rich food sources.
We found that with a constant food source, the
number of sound pulses decreased signifi-
cantly, as the distance to the food source in-
creased. Food quality with a constant sugar
concentration (2.0M) at the artificial food
source, decreased as the distance to the food
source increased. Thus, decline of food quality
and thus in the number of pulses, may explain
the decreased number of new bees at the food
source.
In conclusion, sound signals should have an
effect on recruitment to food sources. More re-
search is needed to know the detailed functions
of these sounds. The sound signals of the
stingless bees can be also designed as a request
for help to discharge the food brought into the
nest as in honey bees (Seeley, 1995). Neverthe-
less, if the sound produced is only to inform
nestmates of the need to be unloaded, then it
takes much lesser effort to just produce a sim-
ple sound rather than vary the sound parame-
ters depending on the quality of the source
being brought in.
ACKNOWLEDGEMENTS
We would like to thank J. Slaa for her statistical
advice. L.K. Biesmeijer, M.J. Sommeijer, M.J.
Eberhard, J. van Veen and anonymous referees are
thanked for their helpful comments on the manu-
script. We are grateful for species identification done
by R. Ayala. The studies were realized as part of the
Research Program of the Regional Stingless
Beekeeping Project (PRAM), which is financed by
the Netherlands Organization for Higher Education
NUFFIC. We are thankful to the Centro de
Investigaciones Apícolas Tropicales (CINAT) of the
Universidad Nacional and to the Vicerrectoria de
Investigación Universidad de Costa Rica.
Résumé – Les sons chez Melipona costaricensis
(Apidae : Meliponini) : effet de la concentration
en sucres et de la distance à la source de nourri-
ture. Les abeilles sans aiguillon (Meliponini) utili-
sent divers signaux pour localiser une source de
nourriture, mais on dispose de peu d’informations
sur le comportement des butineuses et des recrues à
l’intérieur du nid et il n’est pas certain que la qualité
de la source alimentaire (la concentration en sucres)
soit communiquée par des signaux sonores. Diffé-
rentes concentrations en sucres ont été utilisées pour
déterminer l’effet de la qualité d’une source alimen-
taire sur les sons produits par l’abeille sans aiguillon
Melipona costaricensis dans la ruche. De nombreu-
ses composantes des sons étaient positivement cor-
rélées avec la concentration en sucres : la durée du
chant, la durée de toutes les vibrations, la durée des
vibrations longues et la relation entre le son vibré
long et le son vibré court. La durée des vibrations
courtes et de leurs intervalles était significativement
modifiée par la distance à la source alimentaire.
Nous avons trouvé en outre une relation inverse
entre le nombre de vibrations et la distance à la
source alimentaire (P < 0,00238, r = 0,18).
386 I. Aguilar, D. Bruceño
Les signaux sonores peuvent contribuer à recruter
durant la journée des membres de la colonie pour des
ressources alimentaires dont la richesse augmente.
Avec une source alimentaire constante, le nombre de
vibrations sonores diminuait significativement
quand la distance à la source alimentaire augmentait.
La qualité de la nourriture, avec une concentration
en sucres constante de 2,0M à la source alimentaire
artificielle, diminuait quand la distance à la source
alimentaire augmentait. Ainsi, le déclin de la qualité
de la nourriture, et donc le nombre de vibrations,
peut expliquer le nombre réduit d’abeilles nouvelles
à la source alimentaire.
Nos résultats confirment que les ouvrières de M. cos-
taricensis sont capables de distinguer des solutions
sucrées de diverses concentrations. Diverses espèces
du genre Melipona produisent des sons vibrés dont
les durées sont positivement corrélées avec la dis-
tance à la source alimentaire (Esch et al., 1965 ; Nieh
et Roubik, 1998 ; Nieh, 1999 ; Jarau et al., 2000). Par
contre notre étude montre que la variation de la pro-
babilité d’émettre un chant chez M. costaricensis
s’explique par la concentration en sucres de la
source alimentaire (97 %, r
2
= 0,970). Cette corréla-
tion suggère l’existence d’une autre voie par laquelle
sont signalées des sources alimentaires fructueuses.
abeille sans aiguillon / communication sonore /
concentration en sucres / source alimentaire / dis-
tance / Meliponini
Zusammenfassung – Schallerzeugung bei Meli-
pona costaricensis (Apidae: Meliponini): Wir-
kung von Konzentrationen des Zuckers und
Entfernungen der Nektarquellen. Die Wirkung
der Qualität einer Futterquelle auf die Schallerzeu-
gung im Volk wurde bei der stachellosen Biene Me-
lipona costaricensis anhand von unterschiedlichen
Konzentrationen von Zucker untersucht. Verschie-
dene Eigenschaften der erzeugten Töne waren posi-
tiv mit der Zuckerkonzentration korreliert: die
Dauer der Schallerzeugung, die Dauer aller Vibra-
tionspulse, die Dauer von langen Vibrationspulsen
und das Verhältnis von kurzen zu langen Vibrations-
pulsen. Die Zeit zwischen den Schallproduktionen
war mit der Zuckerkonzentration negativ korreliert.
Die Dauer der kurzen Vibrationen und ihren Unter-
brechungen wurde signifikant durch die Entfernung
beeinflusst. Außerdem fanden wir ein inverses Ver-
hältnis zwischen Zahl an Vibrationspulsen und der
Entfernung der Futterquelle (P < 0,00238, r = –0,18).
Schallsignale könnten zur Rekrutierung von Stock-
genossinnen zu immer reichhaltigeren Futterstellen
während des Tages beitragen. Wir fanden, dass bei
konstanter Futterquelle die Anzahl der Schallzei-
chen signifikant abnahm, wenn die Entfernung zum
Futter größer wurde. Die Qualität der Futterquelle
nahm bei konstanter Zuckerkonzentration (2,0M) an
der künstlichen Futterstelle mit zunehmender Entfer-
nung des Futters ab. Demnach könnte die Erreichbar-
keit des Futters und die entsprechende Änderung in
der Anzahl der Vibrationen die abnehmende Zahl von
neuen Bienen am Futter erklären.
Unsere Ergebnisse bestätigen, dass Arbeiterinnen
von M. costaricensis Zuckerlösungen verschiedener
Konzentration unterscheiden können. Mehrere Ar-
ten der Meliponinen erzeugen rhytmische Töne, de-
ren Dauer posi t iv mit der Entfernung der
Futterquellen korreliert sind (Esch et al., 1965; Nieh
und Roubik, 1998; Nieh, 1999; Jarau et al., 2000).
Andererseits zeigt unsere Studie, dass die Variation
in der Wahrscheinlichkeit einer Schallerzeugung bei
M. costaricensis durch die Zuckerkonzentration der
Futterquelle (97 %, r
2
= 0,970) erklärt werden kann.
Diese Korrelation legt einen alternativen Weg zur
Anzeige einer profitablen Futterstellen nahe.
Schallerzeugung / Zuckerkonzentration / Entfer-
nung / stachellose Bienen
REFERENCES
Biesmeijer J.C. (1997) Colony foraging activity of
stingless bees in Costa Rica, with special reference to
pollen, in: The organization of foraging in stingless
bees of the genus Melipona, Thesis, Utrecht University.
Biesmeijer J.C., Richter J., Smeets M., Sommeijer J.M.
(1999) Niche differentiation in nectar-collecting
stingless bees: the influence of morphology, floral
choice and interference competition, Ecol. Entomol.
24, 380–388.
Esch H. (1964) Beiträge zum Problem der
Entfernungsweisung in den Schwänzeltänzen der
Honigbiene, Z. Vgl. Physiol. 48, 534–546.
Esch H. (1967) Die Bedeutung der Lauterzeugung für die
Verständigung der stachellosen Bienen, Z. Vgl.
Physiol. 56, 199–220.
Esch H., Esch I., Kerr W.E. (1965) Sound: an element com-
mon to communication of stingless bees and to dances
of the honeybee, Science 149, 320–321.
Frisch K. von (1967) The dance language and orientation
of bees, Harvard University Press, Cambridge, MA,
USA.
Hrncir M., Jarau S., Zucchi R., Barth F.G. (2000) Recruit-
ment behavior in stingless bees, Melipona scutellaris
and M. quadrifasciata. II. Possible mechanisms of
communication, Apidologie 31, 93–113.
Jarau S., Hrncir M., Zucchi R., Barth F.G. (2000) Recruit-
ment behavior in stingless bees, Melipona scutellaris
and M. quadrifasciata. I. Foraging at food sources dif-
fering in direction and distance, Apidologie 31, 81–91.
Johnson L.K. (1987) Communication of food source loca-
tion by the Stingless bee Trigona fulviventris, in: Eder
J., Rembold H. (Eds.), Chemistry and Biology of So-
cial Insects, Verlag J. Peperny, München, pp. 698–699.
Sounds in Melipona costaricensis 387
Kerr W.E. (1959) Bionomy of Meliponids. VI. Aspects of
food gathering and processing in some stingless bees,
in: Cornell University (Ed.), Symposium on food
Gathering Behavior of Hymenoptera, Ithaca, N.Y.,
pp. 2–4.
Kirchner W.H. (1993) Acoustical communication in hon-
eybees, Apidologie 24, 297–307.
Kirchner W.H., Dreller C. (1993) Acoustical signals in the
dance language of the giant honeybee, Apis dorsata,
Behav. Ecol. Sociobiol. 33, 67–72.
Kirchner W.H., Sommer K. (1992) The dance language of
the honeybee mutant diminutive wings, Behav. Ecol.
Sociobiol. 30, 181–184.
Kirchner W.H., Dreller C., Grasser A., Baidya D. (1996)
The silent dances of the Himalayan honeybee, Apis
laboriosa, Apidologie 27, 331–339.
Kirchner W.H., Lindauer M., Michelsen A. (1988) Honey-
bee Dance Communication. Acoustical indication of
Direction in round Dances, Naturwissenschaften 75,
629–630.
Lindauer M., Kerr W.E. (1958) Die gegenseitige
Verständigung bei den stachellosen Bienen, Z. Vgl.
Physiol. 41, 405–434.
Lindauer M., Kerr W.E. (1960) Communication between
the workers of stingless bees, Bee World 41, 29–71.
Michelsen  A.,  Towne  W.F.,  Kirchner  W.H.,  Kryger  P.
(1987) The acoustic near field of a dancing honeybee, J.
Comp. Physiol. A 161, 633–643.
Michelsen A., Andersen B.B., Kirchner W.H., Lindauer M.
(1989) Honeybees can be recruited by means of a me-
chanical model of a dancing bee, Naturwissenschaften
76, 277–280.
Michelsen A., Andersen B.B., Storm J., Kirchner W.H.,
Lindauer M. (1992) How honeybees perceive commu-
nication dances, studied by means of a mechanical
model, Behav. Ecol. Sociobiol. 30, 43–150.
Nieh J.C. (1998a) The role of a scent beacon in the commu-
nication of food location by the stingless bees,
Melipona panamica, Behav. Ecol. Sociobiol. 43,
47–58.
Nieh J.C. (1998b) The food recruitment dance of the
stingless bee, Melipona panamica, Behav. Ecol.
Sociobiol. 43, 133–145.
Nieh J.C. (1999) Stingless-Bee Communication, Am. Sci.
87, 428–435.
Nieh J.C., Roubik D.W. (1995) A stingless bee (Melipona
panamica) indicates food location without using a
scent trail, Behav. Ecol. Sociobiol. 37, 63–70.
Nieh J.C., Roubik D.W. (1998) Potential mechanisms for
the communication of height and distance by a
stingless bee, Melipona panamica, Behav. Ecol.
Sociobiol. 43, 387–399.
Núñez J.A. (1966) Quantitative Beziehungen zwischen den
Eigenschaften von Futterquellen und dem Verhalten
von Sammelbienen, Z. Vgl. Physiol. 53, 142–164.
Núñez J.A. (1970) The relationship between sugar flow
and foraging and recruiting behaviour of honey bees
(Apis mellifera L.), Anim. Behav. 18, 527–538.
Núñez J.A. (1982) Honeybee foraging strategies at a food
source in relation to its distance from the hive and the
rate of sugar flow, J. Apic. Res. 21, 139–150.
Roubik D.W. (1984) Nectar selection by Melipona and
Apis mellifera (Hymenoptera: Apidae) and the ecology
of nectar intake by bee colonies in a tropical forest,
Oecologia 61, 1–10.
Roubik D.W., Yanega D., Aluja M., Buchmann S.L., Inouye
D.W. (1995) On optimal nectar foraging by some
tropical bees (Hymenoptera: Apidae), Apidologie 26,
197–211.
Seeley T.D. (1986) Social foraging by honeybees: how col-
onies allocate foragers among patches of flowers,
Behav. Ecol. Sociol. 19, 343–354.
Seeley T.D. (1995) The wisdom of the hive, Harvard Uni-
versity Press, Cambridge, MA, USA.
Spangler H.G. (1991) Do honey bees encode distance in-
formation into the wing vibrations of the waggle
dance?, J. Insect Behav. 4, 15–20.
Waddington K.D., Kirchner W.H. (1992) Acoustical and
Behavioral correlates of profitability of food sources in
honey bee round dances, Ethology 92, 1–6.
Wilson E.O. (1971) The insect societies, Harvard Univer-
sity Press, Cambridge, MA, USA.
388 I. Aguilar, D. Briceño