Genetic and morphometric variation in honeybee (Apis mellifera L.) populations of Turkey
Genetic and morphometric variation in honeybee (Apis mellifera L.) populations of Turkey
n KANDEMIR 0
l KENCE 0
Aykut KENCE 0
0 Department of Biology, Middle East Technical University , 06531 Ankara , Turkey
- Six enzyme systems were studied to determine the genetic variability in honeybee populations in Turkey. Ten morphometric characters were also measured to determine the extent of morphometric variation. Out of six enzyme systems, four were found to be polymorphic with 16 allozymes. The average heterozygosity was calculated as 0.072 ± 0.007. Morphometric and electrophoretic variables were equally effective in discriminating honeybee populations. European and Anatolian honeybees were separated on the first axis, and Anatolian honeybees were further separated along a second canonical axis. The observation of rare alleles in isoenzymes, detection of high genetic diversity and the presence of four known subspecies support the argument that Anatolia has been a genetic center for honeybee populations in the Near East.
] claimed that southwest Asia
is a zone of high morphological
diversification and evolution for honeybees. Many
clearly distinct races have evolved within
this region, which includes a diversity of
habitats. Asia Minor, including Anatolia,
appears to be the genetic center for these
honeybee subspecies according to the
multivariate statistical analysis of
morphometric data [
]. Honey bee races in this region
include the subspecies Apis mellifera
anatoliaca, A. m. caucasica, A. m. meda, and
A. m. syriaca, which were considered by
] to form a basal branch (O) of
the species. Another subspecies that is found
in the European part of Turkey, i.e., Thrace,
may be A. m. carnica, which belongs to the
branch C of Ruttner’s classification.
Migratory beekeeping has become
widespread in Turkey within the last
20–30 years. Thousands of colonies are
overwintered in the Mediterranean and
Aegean regions, and then moved to central
and eastern Anatolia during the summer and
fall. These practices might promote the gene
flow between different races, and result in
homogenization of the gene pool of
Despite the apparent importance of
Anatolia in the evolution of honeybees, very
little work has been done on the
morphological and genetic diversity of Anatolian
]. In this study, we aimed
to determine the extent of morphometric
and genetic variation of honeybees
distributed widely across Turkey. Ten
morphometric variables were measured, and
electrophoretic variation was studied in six
2. MATERIALS AND METHODS
Honeybee samples were collected in
1994–1996 between March and September
in Turkey. Samples were taken from 77
different locations in 36 provinces from
different geographic regions of Turkey. Turkey
is divided into seven geographic regions
differing both in climatic conditions and in
geological structure. Sampling was carried
out mostly from small apiaries which do not
practice migratory beekeeping, and the hives
sampled were stationary during the
MarchSeptember sampling period. Requeening of
colonies was mostly natural, although some
beekeepers reported that occasionally queens
had been purchased for some colonies. In
all cases we attempted to sample colonies
that had no history of management for
requeening. Special care was taken to
sample from localities that were not frequented
by migratory beekeepers. Approximately 3
000 worker bees were collected, and were
put into small plastic bottles, which were
labeled; the insects were fed either with
honey cake (honey and powdered sugar
[1:1]) or with ‘Turkish delight’ (water +
saccharose + starch), and brought live to the
laboratory. Honeybees were dissected, the
thoraces were ground, and the homogenates
were kept frozen until needed for
Forewings and hind legs were mounted
on a microscope slide for morphometric
analysis. Microscope slides of legs and
wings were projected onto a TV screen, and
measurements were taken. In the present
study, ten morphometric characters were
measured, i.e., four for the hind legs, four for
the forewings (according to Ruttner [
and an additional two forewing characters,
distance c and distance d as determined by
Six enzyme systems (esterase: 184.108.40.206;
hexokinase: 220.127.116.11; malate dehydrogenase:
18.104.22.168; malic enzyme: 22.214.171.124;
phosphoglucomutase: 126.96.36.199; and phosphoglucose
isomerase: 188.8.131.52), known to be
polymorphic in A. mellifera, were utilized as
biochemical markers. Starch-gel
electrophoresis, gel and sample preparation and
experimental conditions have been reported
]. All allozymes were
designated by using relative mobilities, with the
most common allozyme used as standard
(relative mobility: 100). Gene frequencies,
enzyme heterozygosities and population
heterozygosities were calculated according to
], using BIOSYS [
Goodnessof-fit of genotypic frequencies according to
Hardy–Weinberg expectations were tested
by the c 2-test [
]. Multivariate statistical
analyses were applied to both
morphometric and electrophoretic data using SYN-TAX
]. A phenogram of samples from seven
geographic regions was constructed using
the Mahalanobis distances among centroids
of groups in discriminant function by
UPGMA in NTSYS-PC 1.70 [
Regressions of morphometric and electrophoretic
variables on latitude and longitude were
computed using SYSTAT-7.0 [
3. RESULTS 3.2. Est-3 locus
The mean values of the characters
measured and standard errors have been shown
in Table I, together with the number of hives
and the total number of individuals in each
Analysis of variance (ANOVA) of the
data showed a high heterogeneity among
honeybee populations. Out of 19 variables
(10 morphometric and 9 electrophoretic),
11 displayed significant heterogeneity.
Mdh65 and Mdh100 gene frequencies and
distance d, wing length variables were found
to be highly heterogeneous (P < 0.001).
Out of the six enzyme systems assayed,
four were found to be polymorphic and
two exhibited invariant banding patterns
(Tab. II, and Figs. 1–4). The populations of
honeybees in Turkey were found to be in
Hardy-Weinberg equilibrium with respect
to all polymorphic enzymes, except the Pgm
enzyme system in the majority of southern
honeybee populations, where deviations
were in favor of heterozygotes. Out of
77 sampling localities, in 26 of these, there
were significant deviations in favor of Pgm
heterozygotes (17 localities; P < 0.001;
2 localities; P < 0.01; 7 localities; P < 0.05).
3.1. Pgm locus
The Pgm locus exhibited four alleles,
Pgm45, Pgm63, Pgm75 and Pgm100 according
to their relative mobilities in the present
study. The frequency of the most common
allele (Pgm75) ranged between 0.500–0.976
in 35 polymorphic locations. A significant
linear relationship was revealed by the
regression of the Pgm allele (Pgm75 and
Pgm100) frequencies on latitude. This is the
first report of such a relationship published
in the literature. The distribution of allele
frequencies where the rare alleles are pooled
is given in Figure 1.
The Est-3 locus exhibited three alleles,
Est70, Est100, Est130, as reported previously
for Czechoslovakian honeybees by
Sheppard and McPheron [
], and in central
Anatolian honeybees by Kandemir and Kence
]. These alleles correspond to EstS, EstM,
and EstF respectively, in A. m. ligustica 
and in Greek honeybees . The frequency
of the most common allele at the 22
polymorphic locations ranged between 0.853
and 0.995. Esterase was fixed for the Est100
allele in 14 locations in the northern and
eastern provinces. Generally, there was a
north to south differentiation in esterase
allele frequencies. This conclusion was also
confirmed by a significant linear
relationship between the frequency of Est70 and
latitude. The distribution of allele frequencies
where the rare alleles are pooled is given in
3.3. Hk locus
In the Hk locus, five alleles (Hk77, Hk87,
Hk100, Hk110 and Hk120) were found in
different honeybee populations in Turkey. The
frequency of the most common allele in the
19 locations where this enzyme was
polymorphic ranged between 0.707–0.990. In
Thrace, samples were monomorphic for the
Hk locus. The majority of samples taken
from the Black Sea region were fixed for
the Hk100 allele, whereas hexokinase was
polymorphic in southern Anatolia. There
was also a north to south differentiation in
the hexokinase allele frequencies, shown by
a significant linear relationship between
Hk110 gene frequencies and latitude. The
distribution of allele frequencies where the rare
alleles are pooled is shown in Figure 4.
3.4. Mdh locus
Four alleles (Mdh65, Mdh87, Mdh100 and
Mdh116) were found for this enzyme in the
m .fooN seeb
0 6 2 0 0 6 4 6 3 5 8 3 0 4 4 3 2 9 6 6 4 5 7 1 2 9 2 5 5 3 7 8 1 0 2 3
isn .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00± .00±
2 3 6 2 1 1 1 1 1 1 1 1 1 2 1 2 1 0 2 6 1 1 1 1 1 0 0 0 1 7 5 2 2 3 1 0
icenv ingWitdhw .8250 .9262 .2962 .9200 .0334 .0300 .9295 .0340 .9260 .0303 .9274 .0333 .2794 .0323 .9215 .0308 .9249 .0309 .9277 .3381 .9292 .9234 .9214 .8293 .1304 .0314 .0332 .7252 .2997 .3273 .3029 .3232 .2985 .3055 .3405 .2598
6 7 8 2 8 6 4 3 8 2 2 8 9 0 1 4 3 4 8 8 9 2 2 4 2 5 5 5 1 9 5 4 6 9 7 3
36 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .01 .00 .00 .00 .00 .00 .00 .00 .01 .00 .00 .00 .00 .00
5 2 0 5 2 4 2 3 8 3 7 2 2 7 7 3 2 2 3 1 2 3 0 8 6 2 6 3 3 5 0 2 3 7 1 5
rfom ingWtlegnh .1787 .1987 .7887 .5998 .1029 .0309 .3088 .5609 .0368 .1909 .4588 .5219 .9668 .5129 .6689 .3091 .9998 .7788 .1129 .5239 .4299 .0669 .0019 .1129 .3159 .9918 .1429 .0328 .2319 .0999 .1959 .2259 .2439 .0359 .1349 .0949
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
tiilrrecaeavb itsaecnD D ..816681080± ..813502000± ..190030000± ..815781010± ..914410080± ..915501030± ..814410120± ..916330110± ..818070405± ..915480102± ..914340100± ..024030005± ..814530008± ..919700106± ..186910108± ..195410204± ..198180008± ..148780008± ..169040008± ..271660005± ..250700013± ..169500008± ..169470036± ..158940014± ..280070009± ..159940007± ..189930007± ..016170107± ..519240101± ..519170103± ..619830308± ..719840100± ..619540008± ..919770108± ..319370006± ..219290109±
ohpm ec .000± .000± .000± .000± .000± .000± .000± .000± .200± .000± .000± .000± .000± .000± .000± .000± .000± .000± .000± .100± .000± .000± .100± .000± .000± .000± .000± .010± .000± .010± .010± .000± .000± .000± .000± .040±
rom tisanD C .5508 .4080 .6308 .3908 .0290 .6380 .7080 .8810 .6840 .5800 .6870 .9840 .2860 .9860 .5890 .7890 .8700 .8660 .8800 .1281 .8610 .8570 .8690 .8340 .9260 .8570 .9480 .8330 .8610 .0261 .8840 .7051 .6810 .1980 .9890 .4930
frrrseo1o litaubC B ..05002502± ..00900502± ..22005402± ..00502202± ..00508502± ..00708402± ..00707302± ..00709402± ..10601402± ..00508302± ..00607520± ..00403420± ..00300220± ..00502520± ..00402420± ..00807420± ..00309720± ..00509520± ..00508420± ..00508220± ..00066320± ..00051420± ..02092850± ..00072020± ..00112630± ..00052520± ..00052840± ..00072720± ..00042020± ..00042910± ..00062550± ..00052210± ..00062340± ..00092120± ..00042250± ..00162530±
aand fnoy saaym ltaayn tirvn ilirseak lingö lou iirraaybk irend ilazg iirsseh ay traap irsenM iIrzm rsaK tsaaonuKm irsaeyK ililrraekK aonyK isaanM .rsaaKM laugM inpoS isavS razbonT .rfaSU sakU anV tzaogY lanogduZ raaanKm tiranB raahdnA iIrgd
A A A A A B B B D E E kE taH Is
6 H 8
f K 7
s H 7
z M 0
s G 4
o i e
t lad an n
le .oo iev 13 2 3 22 18 15 7 20 3 15 7 12 36 8 9 4 7 16 7 27 12 16 4 6 6 61 2 2 5 91 4 51 6 5 81 4
present study. The frequency of the most
common allele ranged between 0.637–0.997
in the 18 locations where this enzyme was
polymorphic. Mdh65 is the most frequent of
the rare alleles in Thrace, whereas Mdh116 is
the most frequent rare allele in northeast
Turkey. The Mdh65 allele, which is
infrequent in honeybees in Africa [
]; and the present study), is
common in A. m. ligustica [
] and A. m.
]. The distribution of allele
frequencies where the rare alleles are pooled is
given in Figure 3.
3.5. Pgi and Me loci
Phosphoglucose isomerase (Pgi) and
malic enzyme (Me) were invariant in
Turkish honeybee populations. The Pgi
locus was previously studied in Turkey [
and no genetic variability was detected.
In this study, the heterozygosities of
locations for enzyme loci ranged between
0.012–0.186. Overall average
heterozygosity for Turkish honeybees was calculated as
0.072 ± 0.007. This is the highest mean
heterozygosity reported in A. mellifera to date.
Pamilo et al. [
] and Sylvester [
the mean heterozygosity for European
honeybees as 0.010 and 0.012, respectively.
Later, Sheppard [
] estimated the mean
heterozygosity of A. mellifera from 23
European honeybee colonies as 0.038; he also
noted that the mean heterozygosity was
rather low in other Apis species, except
for A. florea (mean heterozygosities of
A. cerana, A. dorsata, A. florea were 0.004,
0.003 and 0.049, respectively).
3.6. Discriminant function analysis
Honeybees were allocated to the seven
geographic regions of Turkey according to
the geographic position of sampling
locations. A multiple discriminant function
analysis was carried out on the data collected
from the samples taken from the seven
geographic regions by combining the gene
frequencies from electrophoresis and the
measurements of morphometric variables. The
three axes obtained in the multiple
discriminant function analysis explained 87.86%
of the total variation. The proportions of
variation explained by the first, second, and
third axis were 35.54, 32.26, and 20.06%
respectively. Two major groups were
discriminated by the discriminant function
analysis (Fig. 5a). The first group included the
honeybees from Thrace. In this group, the
main variation was along the first
canonical axis. The second group consisted of
Anatolian honeybees (honeybees from the Asian
part of Turkey), which varied mainly along
the second canonical axis. Cubital A, cubital
B, Pgm75, Pgm100, Mdh65, Est70, Est100 were
the variables with the highest loadings on
the first canonical axis, whereas cubital A,
cubital B, distance c, Pgm75, Pgm100, Mdh65,
Mdh100, Est70, and Est100 were loaded highly
on the second canonical axis. In the third
canonical axis, wing length, Pgm75, Pgm100,
Est70, and Est100 were the variables
contributing to the separation of the groups.
When Anatolian honeybees alone were
subjected to discriminant function analysis on
the basis of six geographical regions,
distinct clusters of the sampling provinces,
representing southeastern Anatolia, the
Mediterranean region, central and eastern Anatolia
were formed. Samples from the Aegean
region could not be distinguished from those
in central Anatolia, and samples from the
Black Sea overlapped with the those from
eastern Anatolia to a large extent (Fig. 5b).
Three axes explained 89.68% of the total
variation when honeybee populations of
Thrace were excluded from the analysis.
The variations explained by the three axes
were 47.60, 33.80 and 9.28% respectively.
Cubital A, cubital B, distance c, Pgm75,
Pgm100, Hk110, Mdh65, Mdh100, Est70 and
Est100 were the variables that had the
highest loadings on the first canonical axis.
Cubital A, Pgm75, Pgm100, Hk100, Mdh65
and Mdh100 had high loadings on the
second axis. In the third canonical axis, cubital
A, distance c, Pgm75, Pgm100, Hk100, Mdh65,
Mdh100, Est70 and Est100 were the variables
contributing to the separation of the groups.
When a phenogram of honeybees from
the seven geographic regions was
constructed using Mahalanobis distances, the
Black Sea and east Anatolian samples
clustered very closely. The Aegean and central
Anatolia formed a group, and this group
together with Mediterranean samples made
up a larger cluster. Thrace and southern
Anatolian samples remained as distinct units
within this phenogram (Fig. 6).
When multiple regression analysis was
applied to the morphometric and
electrophoretic variables using latitude and
longitude as independent variables, eight out
of 10 morphometric variables turned out to
be significantly dependent on latitude
(cubital B, distance c, distance d, wing
length, wing width, metatarsus width, femur
and tibia length). Cubital B was also
significantly dependent on longitude.
Among the 12 gene frequencies, six of them
showed a significant relationship with
latitude (Pgm75, Pgm100, Hk110, Mdh65, Mdh100,
and Est70). Mdh65 and Mdh100 were also
significantly dependent on longitude (Tab. III).
Honeybees in Turkey show a high level
of morphometric variation. Of the ten
characters studied, five (distance c, distance d,
wing length, wing width and metatarsus
width) were found to be significantly
different between localities (P < 0.05).
The significant regressions of
morphometric and electrophoretic variables on
latitude and longitude display a structured
pattern in the distribution of populations. The
spatial nature of this pattern is most likely
the result of evolutionary forces acting on
the honeybee populations. This hypothesis
was supported by a spatial autocorrelation
analysis conducted to further determine
relationships among honeybee populations of
]. Morphometric variables that
showed significant regressions on latitude
also had high loadings on the first axis in
the principal component analysis [
axis, known as the size axis [
], allows us
to conclude that the size of the honeybees
increases with increasing latitude. Daly
et al.  showed similar clinal geographic
variation in morphometric characters in feral
colonies of California. Based on the
UPGMA phenogram, the differentiation in
honeybees in Turkey has been maintained,
despite extensive migratory beekeeping.
According to Ruttner [
], A. m.
anatoliaca is distributed throughout central
Anatolia, the Aegean, the Mediterranean, and a
large part of the Black Sea region. A. m.
meda is distributed in southeastern Anatolia,
A. m. caucasica in northeastern Anatolia,
and A. m. carnica in Thrace. This
assessment of subspecies distribution is largely
supported by electrophoretic data and our
morphometric assessments on a reduced
number of characters in honeybee
populations. However, the small set of samples
from southeastern Anatolia form a distinct
cluster that appears to belong to A. m.
syriaca, based on values of CI, wing length, and
body size [
]. Similarly, honeybee
populations in the Mediterranean region, isolated
from the rest of the Anatolian population
by the Taurus mountain range, appear to
form another distinct cluster. Further studies
are needed to determine the taxonomic
status of these honeybee populations by
including additional samples and a full
The honeybees of Turkey were separated
into two groups, the European and the
Anatolian, by discriminant function
analysis. Kirklareli and Edirne honeybee
populations (in Thrace) had the highest Mdh65
gene frequencies, with the highest loadings
on the first canonical axis. We observed that
the Kirklareli honeybee population also had
the highest tibia length. Anatolian
honeybees were separated along the second
canonical axis, with distance c, tibia length, Pgm75
and Pgm100 variables with high loadings.
The Black Sea and east Anatolian samples
had the highest, whereas the Mediterranean
samples had the lowest distance c values,
which separated these two groups on the
Of the 36 provinces from which the
samples were taken, only one province
(K. Maras) was fixed for the Pgm75 allele
based on a very small sample size (12 worker
bees). In all other provinces, Pgm showed a
high degree of polymorphism. Besides
Pgm100 and Pgm75, two additional rare
alleles (Pgm45 and Pgm63) were observed.
There were strong deviations from
Hardy–Weinberg equilibrium in a number of
provinces for Pgm (P < 0.001). Hatay and
S. Urfa samples showed the most extreme
deviations. In Hatay, 269 out of 271
individual honeybees were heterozygous for
Pgm75/100; in S. Urfa, all samples (14
individuals) were heterozygous for the same
alleles. Bingöl, Kirklareli, and Elazig also
showed some degree of deviations in favor
Kandemir and Kence [
] found four
alleles (Hk87, Hk100, Hk110, and Hk120) in
central Anatolian honeybee populations;
and in the present study we found an
additional allele, Hk77, which has not been
reported previously. Out of seven alleles at
the Mdh locus (Mdh55, Mdh65, Mdh80,
Mdh87, Mdh100, Mdh116 and Mdh133),
reported by various authors in different
honeybee populations [
1–3, 5, 6, 12, 16, 17, 21,
25, 26, 32–35
], five of them (Mdh65, Mdh87,
Mdh100, Mdh116, and Mdh133) have been
observed in honeybee populations of Turkey
. The frequency of the Mdh65 allele in
Turkey has been found to be highly reduced;
and in southern and southeastern Anatolia,
the Mdh locus has become invariant
(Mdh100). This relationship has also been
seen in the significant linear regression of
Mdh65 and Mdh100 on latitude and
longitude. This type of clinal variation has been
reported by Nielsen et al. [
] in Mdh
allozymes in Europe, California, and Brazil
with the suggestion that selection may be
involved in many clines. There is some
evidence that fitness differing in Mdh
genotypes may occur, as recent studies have
shown differences in temperature optima
 and differential oxygen consumption
during hovering .
One important result regarding the
electrophoretic analysis is the observation of a
large number of rare alleles (Tab. II). The
existence of rare alleles in a population
suggests that there has not been a recent
bottleneck for the population. The observations
of rare alleles, the presence of four
subspecies and the detection of high genetic
diversity as reflected in the high
heterozygosity support the argument that Anatolia
has been a genetic center for honeybee
populations in the Near East.
This study is the most extensive survey
yet made of the electrophoretic and
morphometric variation in honeybee
populations in the Near East. However, extended
studies including additional morphometric
characters and samples from surrounding
countries (Syria, Iraq, Iran, Georgia and
Armenia) would certainly need to include
a more complete picture of the genetic
variability in the Middle East and Asia.
Résumé – Variation génétique et
morphométrique des populations d’abeilles
domestiques (Apis mellifera L.) en
Turquie. Selon l’analyse statistique
multivariée des données morphométriques, l’Asie
mineure semble être un centre de
diversification génétique pour les races d’abeilles
domestiques qui peuplent cette région, mais
peu d’études ont été consacrées à la diversité
génétique et morphométrique des abeilles
Six systèmes enzymatiques ont été étudiés
pour déterminer la variabilité génétique des
populations d’abeilles en Turquie : enzyme
malique (Me), phosphoglucomutase (Pgm),
estérase-3 (Est), hexokinase (Hk),
phosphoglucose isomérase (Pgi), malate
déshydrogénase (Mdh). Dix caractères
morphométriques ont été mesurés pour déterminer
l’étendue de la variation morphométrique :
huit caractères selon Ruttner [
cubital A et B, longueur et largeur de l’aile
antérieure, longueur du fémur, longueur du
tibia, longueur et largeur du métatarse) et
deux caractères selon Nazzi [
c et d de l’aile antérieure) (Tabs. I et II). La
majorité des variables morphométriques et
enzymatiques ont des relations linéaires
significatives quand on fait une régression
sur la latitude et la longitude (Tab. III).
On a trouvé que quatre des six systèmes
enzymatiques étaient polymorphes et
présentaient 16 isozymes. L’hétérozygosité
moyenne était de 0,072 – 0,007. Les deux
types de données, morphométriques et
électrophorétiques, ont été utilisées pour
discriminer les populations d’abeilles turques.
Les abeilles européennes et les abeilles
d’Anatolie sont discriminées par le 1er axe
canonique, et les abeilles d’Anatolie se
séparent le long du 2e axe. Les variables
morphométriques ont été aussi efficaces que les
variables électrophorétiques pour
discriminer les populations d’abeilles.
Un résultat important concernant l’analyse
électrophorétique est la présence d’un grand
nombre allèles rares (Figs. 1 à 4).
L’existence d’allèles rares dans une population
suggère qu’il n’y a pas eu de goulot
d’étranglement récent. L’observation d’allèles rares,
la présence de quatre sous-espèces connues
et la mise en évidence d’une diversité
génétique élevée démontrée par la forte
hétérozygosité confirment l’argument selon lequel
l’Anatolie a été un centre de diversification
génétique pour les populations d’abeilles
domestiques au Proche-Orient.
Apis mellifera anatoliaca / A. m.
caucasica / A. m. meda / A. m. syriaca /
génétique population / variabilité génétique /
morphométrie / électrophorèse / Turquie
Zusammenfassung – Genetische und
morphometrische Variation in türkischen
Honigbienenpopulationen (Apis melli
fera L.). Zur Bestimmung der genetischen
Variabilität der Honigbienenpopulation in
der Türkei wurden sechs Enzymsysteme
Phosphoglucoseisomerase, Malatdehydrogenase) untersucht.
Zur Bestimmung morphologischer
Variationen wurden zehn morphometrische
Charaktere (Cubitalader A und B, Abstand c
und d, Länge und Breite des Flügels und
des Metatarsus, Länge des Femur und der
Tibia) vermessen. Die meisten
morphometrischen und elektrophoretischen Variablen
zeigten signifikante lineare Beziehungen in
der Regression auf die geographische Breite
und Länge (Cubital B, Abstand c und d,
Flügellänge und -breite, Breite des Metatarsus,
Länge von Femur und Tibia, Pgm75,
Pgm100, Hk110, Mdh65, Mdh100 und Est70).
Bei vier von sechs Enzymsystemen wurden
Polymorphismen mit insgesamt 16
Isoenzymen gefunden. Die mittlere
Heterozygosität wurde zu 0,072 – 0,007 berechnet. Zur
Unterteilung der türkischen Bienen wurden
sowohl die morphometrischen als auch die
elektrophoretischen Daten verwendet.
Europäische und anatolische Honigbienen
waren auf der ersten canonischen Achse
unterschiedlich, die anatolischen Bienen
spalteten sich entlang der zweiten Achse
weiter auf. Morphometrische und
elektrophoretische Variablen waren gleich gut zur
Ein wichtiges Resultat der
elektrophoretischen Analyse war die grob e Zahl seltener
Allele. Dieser Befund legt nahe, dass in der
jüngeren Populationsentwicklung kein
Flaschenhals aufgetreten ist. Der Befund
seltener Allele, die Anwesenheit von vier
bekannten Subspezies, sowie die durch die
hohe Heterozygosität angezeigte grob e
genetische Diversität stützen die
Auffassung, dass Anatolien ein
Entwicklungszentrum der Honigbienenpopulationen des
Nahen Ostens dargestellt hat.
Apis mellifera anatoliaca / A. m.
caucasica / A. m. meda / A. m. syriaca /
Populationengenetik / genetische Variabilität /
Morphometrie / Elektrophorese / Türkei
The authors wish to thank to all who assisted
in the collection, preparation, and measurement
of the samples. Thanks are due to the Ministry of
Agriculture and Rural Affairs of Turkey for
providing assistance during the collection of samples,
and to the beekeepers. This work was supported
by grant VHAG-1077 from the Turkish Scientific
and Technical Research Council to Meral Kence.
Badino G. , Celebrano G. , Manino A. , Population structure and Mdh-1 locus variation in Apis mellifera ligustica , J. Hered . 74 ( 1983 ) 443 - 446 .
Badino G. , Celebrano G. , Manino A. , Longo S. , Enzyme polymorphism in the Sicilian honeybee , Experientia 41 ( 1985 ) 752 - 754 .
Badino G. , Celebrano G. , Manino A. , Ifantidis M.D. , Allozyme variability in Greek honeybees (Apis mellifera L .), Apidologie 19 ( 1988 ) 377 - 386 .
Coelho J.R. , Mitton J.B. , Oxygen consumption during hovering is associated with genetic variation of enzymes in honeybees , Funct. Ecol . 2 ( 1988 ) 141 - 146 .
Contel E.P.B. , Mestriner M.A. , Martins E. , Genetic control and developmental expression of malate dehydrogenase in Apis mellifera , Biochem. Genet . 15 ( 1977 ) 859 - 876 .
Cornuet J.M. , The MDH system in honeybees of Guadeloupe, J . Hered. 70 ( 1979 ) 223 - 224 .
Res. 41 ( 1995 ) 45 - 47 .
Daly H.V. , Hoelmer K. , Gambino P. , Clinal geographic variation in feral honeybees in California , USA, Apidologie 22 ( 1991 ) 591 - 609 .
Del Lama M. A ., Mestriner M.A. , Pavia J.C.A. , Est-5 and Pgm1: new polymorphisms in Apis mellifera , Rev. Brazil. Genet . 8 ( 1985 ) 17 - 27 .
 Del Lama M.A. , Figueiredo R.A. , Soares A.E.E. , Del Lama S.N. , Hexokinase polymorphism in Apis mellifera and its use for Africanized honeybee identification , Rev. Brazil. Genet . 11 ( 1988 ) 287 - 297 .
 Ftayeh A. , Meixner M. , Fuchs S. , Morphometrical investigation in Syrian honeybees, Apidologie 25 ( 1994 ) 396 - 401 .
 Gartside D.F. , Similar allozyme polymorphism in honeybees (Apis mellifera) from different continents , Experientia 36 ( 1980 ) 649 - 650 .
 Jolicouer P. , Mosimann J.E. , Size and shape variation in the painted turtle: a principal component analysis , Growth 24 ( 1960 ) 339 - 354 .
 Kandemir I. , Kence A. , Allozyme variation in a central Anatolian honeybee (Apis mellifera L.) population , Apidologie 26 ( 1995 ) 503 - 510 .
 Kandemir I. , Genetic and morphometric variation in honeybee (Apis mellifera L.) populations in Turkey, PhD diss ., Middle East Tech. Univ. , Ankara, Turkey, 1999 .
 Lobo J.A. , Del Lama M.A. , Mestriner M.A. , Population differentiation and racial admixture in the Africanized honeybee (Apis mellifera L .), Evolution 43 ( 1989 ) 794 - 802 .
 Meixner M.D. , Sheppard W.S. , Dietz A. , Krell R. , Morphological and allozyme variability in honeybees from Kenya, Apidologie 25 ( 1994 ) 188 - 202 .
 Mestriner M.A. , Contel E.P.B. , The P-3 and Est loci in the honeybee Apis mellifera , Genetics 72 ( 1972 ) 733 - 738 .
 Munyan J. , SYSTAT 7 . 0 for Windows, Standard Version, SPSS Inc ., 1997 .
 Nazzi F. , Morphometric analysis of honeybees from an area of racial hybridization in northeastern Italy , Apidologie 23 ( 1992 ) 89 - 96 .
 Ndiritu D.W. , Mutugi N. , Ndung'u S., Variation in malate dehydrogenase allozymes among honeybee populations in Kenya , J. Apic. Res . 25 ( 1986 ) 234 - 237 .
 Nei M. , Molecular Evolutionary Genetics, Columbia Univ. Press, 1972 .
 Nielsen D. , Page Jr P.E. , Crosland M.W.J. , Clinal variation and selection of MDH allozymes in honeybee populations , Experientia 50 ( 1994 ) 867 - 871 .
 Nunamaker R.A. , Wilson W.T., Some isozymes of the honeybee , Isozyme Bull . 13 ( 1980 ) 111 - 112 .
 Nunamaker R.A. , Wilson W.T., Comparison of Mdh allozyme patterns in the African honeybee (Apis mellifera adansonii L.) and the Africanized populations of Brazil, J . Kans. Entomol.
Soc. 54 ( 1981 ) 704 - 710 .
 Nunamaker R.A. , Wilson W.T., Haley B.E. , Electrophoretic detection of Africanized honeybees (Apis mellifera scutellata) in Guatemala and Mexico based on malate dehydrogenase allozyme patterns , J. Kans. Entomol. Soc . 57 ( 1984 ) 622 - 631 .
 Pamilo P. , Varvio-Aho S.L. , Pakkarinen A. , Low enzyme gene variability in Hymenoptera as a consequence of haplodiploidy , Hereditas 88 ( 1978 ) 93 - 99 .
 Podani J. , SYN-TAX-PC Computer Programs for Multivariate Data Analysis in Ecology and Systematics, Version 5 .0, Scientia Publ. , Budapest, 1993 .
 Rohlf F.J. , NTSYS-PC Numerical Taxonomy and Multivariate Analysis System, Version 1 .70, Exeter Software , New York, 1992 .
 Ruttner F. , Biogeography and Taxonomy of Honeybees , Springer-Verlag, Berlin, Heildelberg, 1988 .
 Sheppard W.S. , Genetic variation and differentiation in honeybees (Apis) , Ph.D. diss. , Univ. Illinois at Urbana-Champaign, IL, 1986 .
 Sheppard W.S. , Comparative study of enzyme polymorphism in United States and European honeybee (Hymenoptera: Apidae) populations , Ann. Entomol. Soc. Am . 81 ( 1988 ) 886 - 889 .
 Sheppard W.S. , Berlocher S.H. , Enzyme polymorphism in Apis mellifera from Norway , J. Apic. Res . 23 ( 1984 ) 64 - 69 .
 Sheppard W.S. , Berlocher S.H. , New allozyme variability in Italian honeybees , J. Hered . 76 ( 1985 ) 45 - 48 .
 Sheppard W.S. , McPheron B.A. , Genetic variation in honeybees from an area of racial hybridization in western Czechoslovakia , Apidologie 17 ( 1986 ) 21 - 32 .
 Sheppard W.S. , Soares A.E.E. , DeJong D. , Shimanuki H. , Hybrid status of honeybee populations near the historic origin of Africanization in Brazil , Apidologie 22 ( 1991 ) 643 - 652 .
 Smith D.R. , Slaymaker A. , Palmer M. , Kaftanoglu O. , Turkish honeybees belong to the east Mediterranean lineage , Apidologie 28 ( 1997 ) 269 - 274 .
 Sokal R.R. , Rohlf F.C. , Biometry , Freeman W.H. and Co., San Francisco, CA, 1981 .
 Swafford D.L. , Selander R.B. , BIOSYS-1: a fortran program for the comprehensive analysis of electrophoretic data in population genetics and systematics , J. Hered . 72 ( 1981 ) 281 - 283 .
 Sylvester H.A. , Allozyme variation in honeybees (Apis mellifera L .), Ph.D. diss. , Univ. California, Davis, CA, 1976 .