Journal of Apicultural Research

ISSN: 0021-8839 (Print) 2078-6913 (Online) Journal homepage: http://www.tandfonline.com/loi/tjar20

Observations on the Dance Communication and

Natural Foraging Ranges of Apis Cerana, Apis
Dorsata and Apis Florea in Sri Lanka

R. W. K. Punchihewa, N. Koeniger, P. G. Kevan & R. M. Gadawski

To cite this article: R. W. K. Punchihewa, N. Koeniger, P. G. Kevan & R. M. Gadawski (1985)

Observations on the Dance Communication and Natural Foraging Ranges of Apis Cerana,
Apis Dorsata and Apis Florea in Sri Lanka, Journal of Apicultural Research, 24:3, 168-175, DOI:
10.1080/00218839.1985.11100667

To link to this article: http://dx.doi.org/10.1080/00218839.1985.11100667

Published online: 24 Mar 2015.

Submit your article to this journal

Article views: 4

View related articles

Citing articles: 10 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tjar20

Download by: [Tufts University] Date: 01 December 2016, At: 18:59

Journal of Apicultural Research 24(3): 168-175 (1985)

OBSERVATIONS ON THE DANCE COMMUNICATION AND NATURAL

FORAGING RANGES OF AP/5 CERANA, AP/5 DORSA TA AND
AP/5 FLOREA IN SRI LANKA
1 2 3 3
R. W. K. PuNCHIHEWA , N. KoENIGER , P. G. KEVAN AND R. M. GADAWSKI
1
Apiculture Development Centre, Department of Agriculture, Bindunuwewa, Bandarawela,
2
Sri Lanka
lnstitut fiir Bienenkunde, (Volle Technische Gesellschaft) Fachbereich Biologie, Universitat
Frankfurt!M, Karl-von-Frisch-Weg 2, 6370 Oberursel, German Federal Republic
3
Department of Environmental Biology, University of Guelph, Guelph, Ontario, Canada
NJG 2Wl

Revised manuscript received for publication 1 October 1984

Summary
The dance communication of Asiatic honeybees is qualitatively similar to that of A pis mellifera. However, in A.
cerana and A. fiorea foraging at artificial feeders the dance tempo was observed to decline (i.e., the time per
dance increased) more rapidly than has been reported for A. mellifera as distance to the food source increased.
A pis cerana and A. fiorea forage over short distances; at 3 of the 6 sites in this trial A. cerana did not visit feeders
beyond 500 m. In A. dorsata the change in dance tempo with increasing distance to forage was much slower and
similar, in this respect, to that reported for A. mellifera. Both A. dorsata and A. mellifera are able to forage over
long distances, although in this trial 72% of A. dorsata foragers did not fly beyond 400 m. Apis cerana and A.
fiorea may be unable to communicate long distances to food sources, as these distances would be indicated by
very slow dances. These potential intrinsic limits, and observed constraints on foraging distance in A. cerana,
suggests that beekeepers should manage A. cerana in small, well separated apiaries. In this way floral resources
over small areas would not become so depleted by high densities of bees that honey yields are reduced.

Introduction
The genus Apis is represented by 3 species in Sri Lanka, A. cerana, A. dorsata and A. florea.
Apis cerana and A. florea commonly occur in most parts of the lowland plains throughout the
year. A pis dorsata is common in lowland plains in the early part of the year, but migrates to the
central hills at the onset of the dry season. In the mid-country, all 3 species may be found year
round, although they are not as abundant as in the lowland plains. In the hill country, only A.
cerana is found regularly while A. dorsata occurs seasonally. The cooler climate of the higher
elevations may restrict the distribution of A. florea there.
The largest of the 3 species, A. dorsata, builds a single vertical comb, usually about 0·5 m 2 ,
on the underside of a rock cliff, the eave of a building or a tree branch. Its communication
dance is performed on the vertical comb surface (Lindauer, 1956) and is obvious only when
viewed at nest level. The smallest species, A. §orea, also builds single vertical comb nests
exposed from above, with combs about 0·015 m to 0·025 m 2 • Apisflorea dances on top of the
open comb (Lindauer, 1956; Koeniger et al., 1982) and is readily observed from above. The
nest of A. cerana comprises about 5 or 6 vertical, parallel combs in a protected location such as
a hollow tree trunk or a rock crevice. Lindauer (1956), Michener (1974) and Seeley (1982)
provide further details of the biology of these tropical honeybees and their nests.
Although most climatic factors seem to be favourable for honeybees in Sri Lanka, the honey
production of commercially kept A. cerana is quite low (2-3 kg/hive according to Lanerolle,
1984). Low production may be caused, in part, by an over-concentration of colonies.
Determination of the natural flight range of A. cerana is desirable in order to help recommend
the appropriate distribution of colonies to optimize honey production.
Although our study included comparisons with A. dorsata and A. florea, emphasis was given
to A. cerana as the commercial honey producer. As A. dorsata and A. florea often share the
same habitat with A. cerana, their inclusion facilitated our study of flight ranges.
Flight ranges of honeybees can be determined by behavioural experiments involving
marked bees which have learned to forage for sugar-water (c. 30% sugar) at feeders at known
distances from an experimental colony. At the observation hive or nest, the dance language can
be monitored for individual foragers as they return from the feeder. The speed of the
waggle dance indicates the distance to the food source. This is well described for A. mellifera
 169

by Frisch (1967) who was the first person to relate dance time to distance. Lindauer (1956) also
studied the dance communication of the Sri Lanka honeybees. His data were essential for the
ideas which led to our research.
Materials and Methods
All experiments were conducted in Sri Lanka between December 1980 and July 1981. The
general topography of Sri Lanka, the locations of major cities, and our study sites are shown in
Fig. 1. Information on numbers of colonies used and types of experiments performed at each
site is given in Table 1.
The waggle-dance (Frisch, 1967; Lindauer, 1956) communication pattern was used as the
indicator of flight distance. At the beginning of the study the dance pattern was observed (time
per dance ins) for predetermined distances. As a bee visited the feeding dish it was marked on
the top of the thorax with 'Duco' automobile paint (Smith, 1972). Three to five continuous
and complete dance circuits of marked bees were observed and timed. For any one experiment
there was a feeding dish at only one given distance, therefore we assumed that the marked bees
would indicate only that distance. The possibility that these bees abandoned the dish for other
forage cannot be eliminated, but we feel that this would have been unlikely within the short
period of the experiments. Later, time per dance for each distance was calculated and the
results plotted graphically. In measuring the speed of the waggle dance of the returned

• OVER 900 METRES

D 300-900 METRES

•ANURADHAPURA

eo

70

I I I I
km 50

80°EAST 81° 82°

FrG. 1. Sri Lanka: general topography, major cities and experimental sites.
170
TABLE 1. Locations of experimental sites and numbers of colonies used for obtaining standardization
relationship of dance tempo and distance foraged (SC) or data on natural dances (ND) of honeybees.

No. colonies
A. cerana A. dorsata A. fiorea
District Place sc ND sc ND sc ND

Anuradhapura Puliyankulama 2 2
Anuradhapura (Old Town)
Anuradhapura (New Town) 1* 2

Badulla Bandarawela 3 3
Diyatalawa 2 2(1*)

Colombo Homagama 1*

Kandy Peradeniya 2 1*

Kurunegala Batalagoda 2 1*

*Same colony used for the standard flight curve.

foragers we determined the time per dance rather than the number of dances per 15 s as used
by Frisch (1967) and Lindauer (1956). This made measurements more convenient since it was
unnecessary to judge what fraction of an incomplete dance circuit was performed.
Stop-watches were used to time counted numbers of waggle-dance circuits of foragers
returning from feeders at known distances, or of bees returning from natural foraging trips.
A colony of A. cerana was transferred from its own hive to an observation hive, as described
by Frisch (1967) for A. mellifera except that the observation hive was suitably modified to
accommodate the smaller bee. Experiments were conducted in 6 locations (see Fig. 1),
Puliyankulama (Anuradhapura District), Bandarawela and Diyatalawa (Badulla District),
Homagama (Colombo District), Peradeniya (Kandy District) and Batalagoda (Kurunegala
District). The site at Puliyankulama was a large tract of rice fields, the site at Peradeniya was
an open, predominantly grassland area in the university campus, and the Bandarawela site was
surrounded by tea plantations. We deliberately selected sites with few blossoms or natural
nectar sources to induce A. cerana to fly long distances when foraging from artificial feeders,
thereby providing a more extensive data-set for the standard curve. At a total of 39 feeding
stations, a mean of 34 bees (SD = 19) at each was observed.
A wild colony of A. dorsata was brought to the experimental site at Harischandra, Mawatha,
New Town, Anuradhapura from Nochchiyagama (20 km) on 17 February 1981, using the
technique of Koeniger and Koeniger (1980). At the experimental site the bees quickly became
accustomed to the experimenters and, unless deliberately provoked, were quite docile. It was
possible to begin experimental work within one day of moving the colony. In Peradeniya
natural dances were observed in a colony established on the edge of the roof of a building at the
university. For this experiment a mean number of 58 bees (SD = 5·6) was observed for each
feeding distance.
Two wild colonies of A. florea were brought to our experimental site at Anuradhapura (New
Town, Harischandra, Mawatha) from Nuwara Wewa area on 28 December 1980 and 5
January 1981. The transfer of these colonies was comparatively easy because of their small
size, but had to be done at night to prevent loss of bees from their flying off and absconding
from the open-comb nest. A mean number of 56 bees (SD = 24· 3) was observed for each of 10
distances.
Observations on natural foraging dances
The dances of natural foragers that had returned from their foraging flights were observed as
they entered the colony, and timed as described above. As soon as the dance time of one
forager was recorded, the dance of the next to enter the colony was timed. In this way a
random sample was taken. The three A. cerana colonies used previously for the
standardization procedure were used together with 8 additional colonies to measure the natural
 171
foraging dances. Times per dance of foragers returning from natural sources were calculated
and foraging distance inferred from the standard curve. Bees marked for standardization
experiments were omitted from natural dance observations. Observations were begun after a
lapse of one day from the withdrawal of the artificial feeding dish, to induce them to seek out
natural sources of food. The colonies used for observations on A. dorsata were the same as
those described above. Natural flight ranges were observed and timed for A. florea in the same
way as for A. dorsata and A. cerana. The observations at Anuradhapura and Peradeniya were
made on natural colonies in their natural habitat.

Statistical analysis
Mean, standard deviation and standard error were calculated for the dance time of each species
at each foraging distance. A simple linear-regression analysis was then used to describe the
functional relationship between dance time and foraging distance for each species. Analysis of
covariance was used to test whether the regression equations were identical, and the
Student-Newman-Keuls multiple-range test (q statistic) was employed to determine differ-
ences between slopes and between elevations. The maximum probability of a Type I error was
set at 0·05 for all procedures.

Results
Standardization of dance time-distance curves
Table 2 and Fig. 2 present the standard curve for A. cerana tabulated with the composite data
gathered from 5 experimental sites (see Table 1). We have made a comparison with the data of
Lindauer (1956) with which our results do not agree exactly. Due to lack of methodical
information given by Lindauer (1956) it is difficult to surmise the reasons. Nevertheless, both
curves indicate that short flight-ranges (steep slope: fast decline in dance tempo with
increasing distance to food) are preferred by A. cerana (cf. A. mellifera and A. dorsata,_Fig: 4).

TABLE 2. A summary of the parameters used to describe the simple linear regression equation for each bee
species.
P = probability associated with F; ,-2 = coefficient of determination.
y F
Species II Slope illlercept for H 0 :b=O p ,-2

A. cerana 1326 0·0061 1·3989 20967'28 <0·0001 0·941

A. cerana• 0·0075 1·2192

A. florea 561 0·0065 1·4700 1762·52 <0·0001 0·759

A. florea" 0·0079 1·7260

A. dorsata 230 0·0021 1·6932 194·97 <0·0001 0·461

A. dorsata" 0·0031 1·3266

A. me/liferab 274c 0·0022c 1·3761c

A. melliferab 243d 0·0029d 1·4650d

"From transformed data of Lindauer (1956)

bFrom transformed data of Schweiger (1958)
cFor rapid dancers (young bees)
dFor slow dancers (older bees)

The data of this study (Fig. 3 and Table 2), as well as those of Lindauer (1956), show a
resemblance of A. florea to A. cerana in its preference for short flight ranges (steep slope).
The data on A. dorsata from our study, as well as from that of Lindauer (1956), are
indicative of an ability to forage over long distances (shallow slope: slow decline in dance
tempo with increasing distance to food, Fig. 4). In this characteristic it resembles A. mellifera
which has the ability to forage over long distances (Frisch, 1967).
 172
6. A floreo

(f) 5 (f) 5
~ ~
0
8 4 tl 4 /
/ LWldouer (1956)
~ (f)
/
/
;;; /
uu 3 uu 3 /
u u
z z
("\ 2 ("\
a: y•l39895 •000605x
uu
0. y • 147002 +000649 X
~ I
;:::
0+--.---.--.---.--.--~~ 0+-~------.-~.-----~~
0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
DISTANCE METRES DISTANCE METRES

FIG. 2. Relationship between foraging distance FIG. 3. Relationship between foraging distance
and dance time for A. cerana in Sri Lanka. and dance time for A. florea in Sri Lanka. Solid
Solid line is for present study, broken line for line is present study, broken line is data from
data from Lindauer (1956). Lindauer (1956).

- Lindauer 11956)
5
en --- Ador§OIO.
~ - · - A. mellifero slow dance
u 4' - ··- A mellifera rapid donee
~

~ ~--"'~__:--_:_
3
a 2 - .-:----- ·--- FIG. 4. Relationship between foraging distance
~ -· .~.~~ and dance time for Apis dorsata in Sri Lanka,
~ 1 y•l69319•000213x compared with data of Lindauer (1956) for the
;:::
same species and with slow and rapid dances of
0+--.---.--.---.--.--~~ A. mellifera from Schweiger (1958).
0 100 200 300 400 5CO 600 700
DISTANCE METRES

In Table 2 we show the standings ofthese curves to each other as well as to the curve for A.
mellifera (see also Fig. 4) based on the data of Schweiger (1958) for young and old bees (we
could not distinguish between the ages of the bees). The slope and Y-intercept of the simple
linear regression equation for dance time and distance for each bee species are compared, and
the significance of each regression and the corresponding probability that the calculated slope
could have come from sampling a population with a slope = 0 are also given in Table 2, as is
the coefficient of determination for each regression.
The regression lines for the 3 bee species do not have the same slope (F = 92·4, P<0·0005),
or elevation (F = 1·84 x 105, P<0·005). The slope for A. dorsata is different from the slope
for A. cerana (q = 13·52, P<0·001) and A. florea (q = 9·12, P<0·001); however, A. cerana
and A. florea have slopes which are not significantly different (q = 1·64, P>0·20). The
elevation of the line for A. cerana is significantly different from the elevation of the line for A.
florea (q = 25·00, P<0·001) and A. dorsata (q = 17·55, P<0·001); however, the elevations of
the lines for A. florea and A. dorsata are not significantly different (q = 0·05, P>0·50).
Dance time and natural foraging distance
The time per dance of returning foragers from natural sources was calculated as described
above. These times were then categorized into different time intervals of 0·5 sand the results
used to construct histograms showing the frequency for each dance time. These histograms
were then fitted with the standard foraging-distance regression curves (Fig. 5 to Fig. 7). This
 173

500

(f) ~ ~~ florea
z \i:400 400~
0
~2000 ~ >-
w w
ll:
~300 ::;;
til
(!] lL
;;;
0 0 w
lL a:200
w 200~
0
ll:
w
§! 1000
~
iii 100 ~
::0
z
0+---1-~1-~-r_L-+_L-+~~~~----ro
0 LO 20 30 40 50 60 70 80
TIME PER DANCE CATEGORIES IN SECONDS

0 10 20 30 40 50 60 70 80
TIME PER DANCE CATEGORIES IN SECOI1bS

40 BOO

o-~ I (f)
I I I w
30 I I 60011:
(f)500 ~R§ dorsato I
>--
z w
0 ::;;
(f)
~4 ;;;
A cerono
400(f) z
ll: w g /A floreo w
w u
(f)
ll:
>-- \i _._
z
~ 300 ~20 400;:!
~ w
(f)
(f)
"- ;;; (!] 0
0 0
0: 200 200i:'J ;!i
w z
CD
::;; ;'!
(f)
::> 10 200
z 100 i5

0
0 10 20 30 40 50 60 7D 80
TIME PER DANCE CATEGORIES IN SECONDS t---,_
0
0 10 20 30 40 50 60 70 80
TIME PER DANCE IN SECONDS

FIG. 5 (upper left). Natural foraging ranges for A. cerana in Sri Lanka shown by the frequency of dance-time
categories (bar graph) and the corresponding distances represented (oblique line).
Data from Fig. 2.

FIG. 6 (upper right). Natural foraging ranges for A. fiorea in Sri Lanka shown by the frequency dance-time
categories (bar graph) and the corresponding distances represented (oblique line).
Data from Fig. 3.

FIG. 7 (lower left). Natural foraging ranges for A. dorsata in Sri Lanka shown by the frequency of dance-time
categories (bar graph) and the corresponding distances represented (oblique line).
Data from Fig. 4.

FIG. 8 (lower right). Comparisons of natural foraging ranges of the three Apis species in Sri Lanka.
(---),A. cerana; (---),A. fiorea; (----),A. dorsata. Frequency of a dance-time category is presented as
percent of observations (data from Fig. 5, Fig. 6 and Fig. 7) and corresponding distance represented (oblique
lines from Fig. 2, Fig. 3 and Fig. 4). Arrows show cumulative percentage (to 70%) of the populations of
foragers for each species.
174
allowed interpolation of the distance that the bees in each category foraged from each
dance-time category.
For interspecific comparisons we have combined in Fig. 8 all the findings presented in Fig.
5, Fig. 6 and Fig. 7. To facilitate this combination we have transformed the dance-time
category for each species into a percentage category of the total number of observations. This
transformation also allowed us to compare the behaviour of a comparable group of foragers in
each species. In this comparison we have taken about 70% (68·3% to 72·5%) ofthe population
studied for each species of bee to compare their foraging radii. For A. cerana (Fig. 5) we have
used categories 1-3 (i.e., the percentage of bees with flight ranges indicated by waggle-dance
circuit times of 1·0-2·5 s inclusive), which represents 68·3% of the foraging bees. For A.
florea, categories 1-6 (i.e., the percentage of bees with flight ranges corresponding to dance
circuit times of 1·0-4·0 s inclusive) represent 72·3% of the foraging bees. For A. dorsata (Fig.
7), categories 1-3 (i.e., the percentage of bees with flight ranges corresponding to
waggle-dance circuit times of 1·0-2·5 s inclusive), represent 72·5% of the foraging bees.

Discussion
All three species of Apis have preferred short foraging distances of not more than 400 m
(Fig. 8). The comparative flight curves (Fig. 4) show that A. dorsata and A. mellifera are
similar. Though the shallow standard flight-distance curve of A. dorsata indicates the ability to
forage as widely as A. mellifera, it seemed to prefer to forage close to the nest (Fig. 7); 72% of
the foragers did not fly beyond 400 m. This may explain, in part, our inability to train these ·
bees to a feeding dish beyond 250m (Fig. 4). Presumably, there was plenty offorage in the
vicinity of most experimental sites to which the bees went in preference to the more distant
feeding dish.
There is also a similarity between A. cerana and A. florea. Both foraged within a small radius
(Fig. 5 and Fig. 6). The standard flight-distance curves suggest that they are unable to
communicate longer distances to food sources. These bees perform a relatively longer dance
for the distance they travel than do A. dorsata and A. mellifera; the slope of the line is steeper
(Fig. 8). Only at 3 of the 6 experimental sites were we able to entice A. cerana foragers to visit
feeding dishes beyond 500 m. These sites (Puliyankulama, Peradeniya and Bandarawela) were
especially chosen as they were devoid of the mixed floral types and dense vegetation typically
found in tropical countries (see above).
The dispersion in data points in Fig. 2, Fig. 3 and Fig. 4 may be due to the age differences
of the marked bees trained to the feeding dish (cf. Schweiger, 1958) or to other factors such as
individual colony differences, weather, topography, etc. which are known to affect the dance
language (Frisch, 1967), but which we did not or could not control.
During the early part of the year, after the monsoonal rains, the lowland plains (e.g.,
Anuradhapura District) offer abundant food for all3 species of bees. Hence they would be able
to fulfil their food requirements within a short distance. Probably the large A. dorsata and
medium-size A. cerana, with large colony populations, are able to exploit all the large flowers
in an area. Within the same area, smaller A. florea may exploit the flowers that the larger bees
do not reach, or which are undesirable because of size or anatomical differences in flowers, or
bees, or both. This combination of phenomena may enable the 3 species to coexist. However,
with the approach of the dearth period, the smaller A. florea and A. cerana appear to become
more successful competitors for nectar than A. dorsata. Then the option left for the larger A.
dorsata is to make a long-distance migration to an altogether different environment. The
experiments of Koeniger and Vorwohl (1979) on the competition among Apini further
strengthen this point; they found that A. dorsata was a poorer competitor than other, smaller
social bees feeding at dishes. Once A. dorsata removes itself from local foraging niches, A.
cerana and A. florea may coexist through a system of resource partitioning. From the results of
pollen analyses of natural honeys, we suspect that the different bees forage at different types of
flowers. Some data supporting this are given by Koeniger and Vorwohl (1979), but certainly
this is an area for important future research.
The trends we have described agree in general with those of Lindauer (1956, 1957). He
suggests that the short foraging ranges of these bees are in response to rain, the abundance of
tropical flowers, lack of climatic seasonality, and a weaker hoarding instinct than in A.
mellifera. We are doubtful of the importance of rain because it is not rainy year round in Sri
 175
Lanka and even during the rains there is ample warning for bees to return to the nest from a
distance of over 1 to 2 km before being washed out. The abundance of tropical flowers
certainly will restrict observed foraging ranges and is probably important for A. florea and A.
cerana, but not for A. dorsata. The lack of climatic seasonality may account for the bees'
reduced requirements for hoarding, and so effect a lesser need for extensive foraging ranges.
These ideas are, in part, paralleled by suggestions of the reasons for differences in dialect
among races of A. mellifera (Gould, 1982).
The demonstrated short foraging range of A. cerana indicates that large bee-yards are not
appropriate for this species. Large numbers of bees in small areas will deplete floral resources
rapidly and result in severe interspecific competition and low honey yields and high rates of
absconding, as observed. Thus we suggest that beekeepers using A. cerana should spread their
hives out with only a few at each location.
Acknowledgements
We wish to thank the Sri Lanka Department of Agriculture for providing us with the facilities
for this study. We are indebted toR. M. Ratnayake, D. C. Hindurangala, P. P. Premaratna
and R. A. Sellaperuma of the Apiculture Development Centre at Banderawela for their
assistance during this study, and are especially grateful to R. Sellaperuma who was essential to
our finding colonies of A. dorsata and A. florea. R. Longair helped in the preparation of the
manuscript.
References
BuTLER, C. G. (1953) The present status of beekeeping in Ceylon and possibilities for its future development.
Br. agric. Bull. 6 : 125-128
---(1975) The honey bee colony life history. pp. 39-74 from The hive and the honey bee. eds Dadant and
Sons. Hamilton, IL, USA : Dadant & Sons
FRISCH, K. voN (1976) The dance language and orientation of bees. Cambridge, MA, USA : Belknap Press of
Harvard University Press
GouLD, J. L. (1982) Why do honey bees have dialects? Behavl Ecol. Sociobiol. 10 : 53-56
KoENIGER, N.; KoENIGER, G.; PuNCHIHEWA, R. K. W.; FABRITms, M. 0.; FABRITms, M. I. (1982)
Observations and experiments on dance communication in Apis florea in Sri Lanka. J. apic. Res.
21 : 45-52
KoENIGER, N.; VoRWOHL, G. ( 1979) Competition for food among four sympatric species of Apini in Sri Lanka
(Apis dorsata, A. cerana, A. florea, Trigona iridipennis). J. apic. Res. 18(2) : 95-109
LANEROLLE, G. A. (1984) Beekeeping in Sri Lanka. pp. 217-221 from Proceedings of the expert consultation on
beekeeping with Apis mellifera in tropical and subtropical Asia. Rome: FAO .
LINDAUER, M. (1956) Uber die Verstandigung bei indischen Bienen. Z. vergl. Physiol. 28 : 521-557
-(1957) Communication among the honeybees and stingless bees in India. Bee Wid 38 : 3-14,34-39
MICHENER, C. D. (1974) The social behavior of the bees. Cambridge, MA, USA.: Belknap Press of Harvard
University Press
ScHWEIGER, E. M. (1958) Uber individuelle Unterschiede in der Entfernungs und Richtungsangabe bei den
Tanzen der Bienen, Z. vergl. Physiol. 41 : 272-299
SEELEY, T. D.; SEELEY, R. H.; AKRATANAKUL, P. (1982) Colony defense strategies of the honeybees in
Thailand. Ecol. Monogr. 52 : 43-63
SMITH, M. V. (1972) Marking bees and queens. Bee Wid 53 : 9-13

K77