Albanian j. agric. sci. 2014;(Special edition)

Agricultural University of Tirana

(Open Access)

RESEARCH ARTICLE

Study on morphometric traits of the Albanian bees MANJOLA KULIÇI1*, KRISTAQ KUME2 1

Internacional University of Tirana, Tirana, Albania

2

Aleksander Moisiu University, Durrës, Albania

*Coresponding author Email: [email protected]

Abstract 19 morphological features, measured on 3600 worker bee samples collected from twenty different regions of Albania were used for the identification and characterization the Albanian bee. The measurement was done using Scan Photo technique (SPT) and Photoshop program, in accordance with the standard method. The average values of the measured parameters were: Proboscis length (PL) (6.55mm), Forewing length (FWL) (9.32mm), Forewing width (FWW) (3.195mm), Hind wing length (HWL) (6.481mm), Hind wing width (HWW) (1.847mm), Femur length (FL) (2.575mm), Tibia length (TL) (3.212mm), Basitarsus length (BL) (2.094mm),Basitarsus width (BW) (1.217mm), Number of hooks (HA) (20.558), Sternite 3 longitudinal (LS3) (2.776mm), Wax mirror of sternite 3 longitudinal (WL) (1.344mm), Wax mirror of sternite 3 transversal (WT) (2.390mm), Distance between wax mirrors of sternite 3(WD) (0.321mm), Tergitite 4 longitudinal (T4) (2.092mm), Tomentum (TOM A) (0.782mm), Width of the dark stripe between tomentum and posterior rim of tergite 4 (TOM B) (0.507mm), Length of hairs on tergite 5 (HLT5) (0.290mm) and Cubital index (CI) (2.779mm). Referring to the values of coefficient of variations, features can be classified into two groups: (i) PL, FWL, FWW, HWL, HWW, TL, BL, BW, HA, LS3, WL, WT, T4 features with low level of variation (2.1% - 6.0%); (ii) FL, WD, TOM A, TOM B, HLT5 features with a high level of variation (12.2% - 36.7%). CI is feature with moderate variation (15.3%). The observed values of the above parameters show that, although during the last twenty years the genes migration processes in the Albanian bee population have been uncontrolled, their classification into the group of Apis mellifera carnica, continues to be a significant hypothesis. Keywords - Honeybee, Apis mellifera, Morphological features, Scan Photo method.

honeybees spread over 20 districts of Albania, that is performed for the first time in a large number of bee families and with a contemporary method SPT [2]. So the aim of this study was the quantitative evaluation of morphological characters, for bees in Albanian population, spread over 20 districts of Albania. By using this evaluation we studied the relation that exist between these morphological traits, and were evaluated affinities/distances between subpopulations of bees, identified as such, by the region where they are reared. With the help of a semi-automated system we aim to increase the efficiency of morphometric analyses by decreasing the time spent on handling these measurements and by increasing the accuracy of measuring hindleg, forewing, hindwing, proboscis, sternite 3, tergite 4 and tergite 5 features. The honeybee samples necessary for the studies that are collected from different areas of Albania were used to prepare slides with the dissected parts of worker bees and morphometric measurements are performed by using 4 length traits from hindleg, 4 length traits from third sternite, 4 length traits from forewings and hind wings, 3 length traits from fourth tergite, length of proboscis, length of hairs in fifth tergite, cubital index and number of hooks.

Introduction The world as we know does not exist, if bees would not be, they are the most economically valuable pollinators of agricultural crops worldwide [7], they perform pollination of about 80% of flower plants. The honeybees of the world are widespread over their native range of Europe, Africa and Asia, and have been successfully transplanted to other regions and continents [3]. Physiological, biological, genetic, and morphological studies have been conducted on this insect. Description and identification of subspecies are usually based on morphological traits, especially measurable ones [3]. In this study morphometric measurements are performed by using 19 morphological features, measured on worker bee samples collected from different areas in Albania. In our country a number of morphological studies were conducted for identification of the race of our bees. Studies made by Nuri [4] and another group of authors [5], identify our bees as variety of carnica race. Later investigation made by Dedej [6], for traits of forewings of honeybees from 8 districs of Albania, defined our bees as an hybrid between A.m. carnica and A. macedonica. We are focused in morphological measurement of 419

Kuliçi & Kume

Materials and methods

Sample collections

Study area

Sampling was done in April-June period 2012 and 2013. For investigating the honeybee populations from different regions of Albania in order to determine variation of populations in these regions the honeybee samples are collected from 20 different districts. The sampling was done in stable apiaries, with satisfactory health status that did not practice queen replacement. In each district are taken samples from three different colonies and in each colony we have taken three samples (from three different bee families). In total we have taken 180 samples of 100 bees each. The honeybee workers collected from each colony are put in separate glass jars. Young worker bees were collected from inside the hive, than were killed and preserved in 70% ethanol [1]. Jars are labelled and information about each colony is recorded. Twenty worker honeybees were selected randomly from each sample and dissected for measurements [1]. Worker bees were dissected by using forceps to separate forewing, hind wing, proboscis, hind leg, tergite 4, tergite 5 and sternite 3, after that, these body parts were mounted on slides. From each worker bee with her body parts we have prepared seven types of permanent microscopic preparations, which were then scanned one by one and used for measurement. For facilitating measurements mounting of body parts was done in the same direction. In total 3600 worker bees from 60 colonies spread over different districts in Albania are used for morphometric analysis..

Morphological character of honeybee in our country, were measured in samples collected from 60 different honeybee colonies spread over 20 different districts in Albania, as shown in figure 1. In each districts are taken samples from three different parks, so in total 60 colonies were sampled.

Figure 1: Albanian map, that shows districts where samples were taken.

Table 1: Morphological characters measured in this study Characters of the head and abdomen

Characters of limbs articulated in the thorax

1. Length of proboscis

( PL)

10. Forewing Length

(FWL)

2. Longitudinal diameter of tergitite 4

(T4)

11. Forewing Width

(FWW)

3. Width of Tomentum, tergite 4

(TOM A)

12. Hind wing length

(HWL)

4.Width of the dark stripe between tomentum and posterior rim of tergite 4

(TOM B)

13. Hind wing width

(HWW)

5. Length of hairs on tergite 5

(HLT 5)

14. Femur Length

(FL)

6. Sternite 3, Longitudinal

(LS3)

15. Tibia Length

(TL)

7. Wax mirror of sternite 3 longitudinal

(WL)

16. Basitarsus Length

(BL)

8. Wax mirror of sternite 3 transversal

(WT)

17. Basitarsus Width

(BW)

9. Distance between wax mirrors St. 3

(WD)

18. Number of hooks

(HA)

19. Cubital index

(CI)

420

Study on morphometric traits of the Albanian bees

Morphological characters green blue (RGB), resolution 4800 ppi and units of mm), a laptop with screen resolution (1440× 900pixels), a computer program (Photoshop) and a ruler are used to measure the features. The slides of dissected parts of honeybees are placed in scanner, which transmit the image to the computer. Images are open through computer software and after this process the morphometric measurements are carried out with measuring tools of Photoshop program installed in the computer. However, performing measurements in this way make it possible simplification of the procedure of measurements, increase the number of features that can be measured and saves in time. Preparations of honeybee forewings, hind wings, hindlegs, tergite 4, tergite 5, sternite 3 and proboscis are registered in separate images as shown in Figure 2.

In table 1 are stated morphological character that are taken into consideration for taxonomic analysis, which are measured by the SPT method. 19 morphological characters were selected from those studied by investigators of subspecific variation [8, 9, 10, 11]. 19 measurements were conducted on each bee approximately 108.000 morphometric measurements were taken for this study. These traits are divided into characters of the head, abdomen, and into characters of limbs articulated in the thorax as are shown in Table 1 Measuring of morphological characters Computer measurement of morphological traits, were conducted with Scan Photo methods with Photoshop software [2]. In Biology Department, a system made of a scanner (Zoom 100%, color is red

Figure 2: Sternite 3 images

Data analysis Sistematics is used based on the statistics. In order to compare one trait, many samples of one region can be measured and statistical analysis is done on obtained results from these samples. The basic information that was subjected to statistical treatment, represented by 180 samples from 20 districts of the country for 19 characters taken in the study. Statgraphic Centurion IX software was used to make statistical analysis of data. Decsriptive analysis was used to estimate the means and standard errors for all morphological traits, object of this study, for every bee subpopulation, identified as such with the geographical area in which it is located.

Results and discussion In table 2 are given evaluations for means of morphological characters taken in the study, measured in honey bee populations from different regions of Albania. Evaluations for means, standard errors and coefficient of variation for the whole population of bees in Albania are given in Table 3. In table 4 are given values of correlation coefficients between different morphological characters for the whole population of the country for 19 characters taken in the study.

421

PL

6.48

6.549

6.633

6.582

6.569

6.417

6.361

6.398

6.229

6.325

6.404

6.379

6.524

6.483

6.544

6.555

6.419

6.419

6.426

6.741

Region

Berat

Erseka

Kruja

Permet

Puka

Tirana

Vlora

Dibra

Elbasan

Fier

Gramsh

Korca

Shkodra

Tropoje

Lezha

Kukes

Librazhd

Lushnje

Mat

Saranda

9.434

9.343

9.152

9.398

9.348

9.223

9.325

9.169

9.392

9.391

9.178

9.441

9.406

9.123

9.172

9.335

9.215

9.443

9.457

9.36

FWL

3.193

3.175

3.189

3.225

3.209

3.179

3.088

3.195

3.232

3.262

3.129

3.307

3.167

3.121

3.159

3.2

3.26

3.233

3.213

3.188

FWW

6.512

6.529

6.428

6.433

6.561

6.438

6.397

6.38

6.514

6.511

6.372

6.555

6.628

6.339

6.374

6.528

6.494

6.567

6.547

6.46

HWL

1.866

1.858

1.884

1.852

1.889

1.832

1.803

1.828

1.835

1.837

1.818

1.846

1.857

1.81

1.838

1.873

1.847

1.878

1.867

1.87

HWW

2.653

2.583

2.628

2.389

2.566

2.59

2.34

2.561

2.636

2.609

2.591

2.637

2.434

2.589

2.529

2.613

2.602

2.671

2.652

2.63

FL

3.281

3.171

3.18

3.239

3.236

3.192

3.209

3.156

3.232

3.194

3.202

3.227

3.217

3.157

3.217

3.197

3.254

3.224

3.223

3.24

TL

2.146

2.122

2.11

1.996

2.097

2.077

2.026

2.086

2.098

2.079

2.061

2.144

2.073

2.108

2.066

2.063

2.13

2.124

2.133

2.13

BL

1.226

1.209

1.201

1.227

1.219

1.201

1.246

1.198

1.216

1.219

1.2

1.244

1.231

1.207

1.217

1.221

1.212

1.214

1.228

1.21

BW

21.6

20.433

20.016

20.902

20.016

20.75

19.644

20.617

21.094

21.066

20.033

20.64

20.394

20.061

20.006

19.822

20.4

20.717

20.728

21.07

HA

2.843

2.743

2.737

2.787

2.798

2.739

2.757

2.753

2.786

2.811

2.697

2.82

2.8

2.725

2.772

2.772

2.807

2.774

2.801

2.80

LS3

1.372

1.322

1.308

1.346

1.342

1.353

1.353

1.314

1.363

1.371

1.333

1.338

1.386

1.298

1.347

1.358

1.348

1.344

1.354

1.34

WL

Table 2. Evaluations of means for morphological characters in bee populations that reared in different regions

2.464

2.356

2.366

2.363

2.386

2.388

2.339

2.373

2.416

2.418

2.381

2.431

2.396

2.342

2.366

2.377

2.404

2.423

2.421

2.39

WT

0.313

0.294

0.336

0.337

0.333

0.32

0.303

0.312

0.383

0.314

0.323

0.314

0.288

0.327

0.338

0.316

0.304

0.331

0.327

0.32

WD

0.835

0.75

0.82

0.784

0.759

0.753

0.749

0.773

0.796

0.783

0.81

0.801

0.753

0.83

0.81

0.754

0.785

0.848

0.832

0.768

TOM

0.318

0.286

0.307

0.353

0.298

0.297

0.267

0.293

0.339

0.317

0.295

0.301

0.274

0.301

0.287

0.276

0.291

0.296

0.299

0.342

HLT 5

2.94

2.648

2.711

2.829

2.769

2.818

2.332

2.664

2.939

2.722

2.757

2.848

2.731

2.731

2.793

2.686

2.887

2.648

2.954

2.896

CI

0.007

6.076

St.err

CV%

2.146

0.001

9.32

FWL

3.004

0.002

3.195

FWW

2.746

0.003

6.481

HWL

4.006

0.001

1.847

HWW

15.262

0.007

2.575

FL

4.261

0.002

3.212

TL

4.345

0.002

2.094

BL

3.944

0.001

1.217

BW

1

0.18

0.00

0.21

0.42

0.20

0.38

0.21

-0.08

0.25

0.31

0.27

0.33

-0.13

0.08

-0.08

0.07

PL

FWL

FWW

HWL

HWW

FL

TL

BL

BW

HA

LS3

WL

WT

WD

TOM A

HLT 5

CI

PL

0.21

0.25

0.00

-0.02

0.62

0.64

0.71

0.53

0.67

0.17

0.57

0.06

0.42

0.79

0.48

1.00

FWL

0.52

0.38

0.14

0.16

0.65

0.22

0.64

0.51

0.16

0.40

0.37

0.46

0.43

0.56

1.00

FWW

0.23

-0.06

-0.12

-0.17

0.61

0.59

0.64

0.30

0.42

0.38

0.43

0.23

0.64

1.00

HWL

0.32

0.16

0.10

0.05

0.37

0.13

0.42

0.18

0.00

0.40

0.36

0.39

1.00

HWW

0.54

0.20

0.50

0.27

0.58

-0.16

0.13

0.37

-0.41

0.80

0.06

1.00

FL

0.51

0.36

0.18

0.14

0.64

0.61

0.74

0.46

0.46

0.19

1.00

TL

0.46

0.04

0.42

-0.06

0.58

-0.15

0.34

0.37

-0.13

1.00

BL

-0.13

-0.12

-0.10

-0.19

0.23

0.53

0.59

0.01

1.00

BW

Table 4. Evaluation of correlation coefficients between morphological traits

6.55

Mean

PL

0.62

0.69

0.25

0.17

0.72

0.37

0.58

1.00

HA

7.987

0.027

20.558

HA

0.46

0.29

0.05

-0.06

0.72

0.63

1.00

LS3

3.206

0.001

2.776

LS3

0.20

0.01

-0.21

-0.09

0.55

1.00

WL

6.027

0.001

1.344

WL

0.61

0.27

0.41

0.09

1.00

WT

3.431

0.001

2.39

WT

0.38

0.57

0.40

1.00

WD

24.456

0.001

0.321

WD

Table 3. Evaluation of means, standard errors and coeficent of variacions for morphological traits in Albanian bee population

0.36

0.25

1.00

TOMA

3.728

0.001

2.092

T4

0.59

1.00

HLT 5

11.125

0.001

0.782

TOMA

1

CI

18.145

0.002

0.507

TOMB

23.793

0.001

00.29

HLT 5

15.329

0.007

2.779

CI

Kuliçi & Kume

From comparison of averages of these morphological traits with those referred by other authors 20 years ago, is noticed that the differences between them are small, statistically non-significant (P>0.05). This result gives us the right to raise the hypothesis according to which, in Albanian bee population, regardless of migratory phenomenons, that have been often uncontrolled and not in implementation of any genetic and racial program, during this period does not occurred significant changes (P <0:05). Referring to values of coefficient of variation, features can be classified into two groups • PL, FWL, FWW, HWL, HWW, TL, BL, BW, HA, LS3, WL, WT, LD4 features with low level of variation (2.1% - 6.0%); • FL, WD, TOM A, TOM B, HLT 5, features with high level of variation (12.2% - 36.7%). CI is a feature with moderate variation (15.3%). To judge relations that may exist between different traits, the Pierson correlation coefficients were evaluated (Table 4). As seen, our data do not confirm any relation of length of proboscis with FWW, TOM A and CI (P>0/05), while this character has a negative relation

(P<0.05) with BW, WD, HLT 5 and positive relation (P<0.05) with HWW, TL, LS3 and WT. Stronger relations (r>0.6) were found between features FWLxHWL, FWLxBW, FWLxLS3, FWLxWL, FWLxWT, FWWxLS3, FWWxWT, HWLxHWW, HWLxLS3, HWLxWT, TLxLS3, TLxWL, TLxWT, HA xWT, HA xHLT 5, HA xCI, LS3 xWL, LS3 xWT, WT xCI. Using average data, evaluated for morphological traits taken in the study, in bee subpopulations that reared in different regions of Albania, was conducted Principal Component Analysis. It resulted that the two first principal components explain about 73.4% of the general variance. Consequently, the estimates for Euclidean distances between the corresponding centroids of subpopulation, identified as such by the region where they are reared, can serve as information to judge about affinities between these subpopulations, groups that can be made and, consequently also for the level of local differentiation in the Albanian bee population. Figure no. 3 shows dendrogram, built by using Euclidean distances between the corresponding centroids of bee subpopulations, in plan of two first principal components.

4

Distance

3

2

1

Tropoje

Librazhd

Korça

Elbasani

Fieri

Tirana

Mat

Dibra

Shkoder

Vlora

Lezhë

Puka

Kukës

Lushnjë

Permeti

Gramshi

Kruja

Erseka

Berat

0

Figure 3. Dendrogram that present groupings of bee subpopulations

The Cluster analysis (Figure 3) shows the existence of a local differentiation between different bee subpopulations in Albania. Grouping of subpopulations is done in some degree. I. The two main groups are: Tropoja; Berat, Erseka, Kruja, Gramsh, Permet, Lushnje, Puka, Kukes, Vlora, Lezha, Shkodra, Diber, Mat, Tirana, Fier, Elbasan, Korca, Librazhd. II. Within the main group you can distinguish 2 other groups close to each other, which are:

III.

IV. 424

Librazhd; Berat, Erseka, Kruja, Gramsh, Permet, Lushnje, Puka, Kukes, Vlora, Lezha, Shkodra, Diber, Mat, Tirana, Fier, Elbasan, Korca. Within the second above group, you can distinguish two other groups: Korca; Berat, Erseka, Kruja, Gramsh, Permet, Lushnje, Puka, Kukes, Vlora, Lezha, Shkodra, Diber, Mat, Tirana, Fier, Elbasan. In continuation of the large group that we defined above, you can see other two related

Study on morphometric traits of the Albanian bees

groups which are: Elbasan; Berat, Erseka, Kruja, Gramsh, Permet, Lushnje, Puka, Kukes, Vlora, Lezha, Shkodra, Diber, Mat, Tirana and Fier. The latter grouping is divided into two main groups: Berat, Erseka, Kruja, Gramsh and Permet, Lushnje, Puka, Kukes, Vlora, Lezha, Shkodra, Diber, Mat, Tirana. If we compare these groups with geographic area where respective bee populations are found, you can notice only a tendency of differentiation phenomenon due to the isolation in distance. In favor of this hypothesis are for example the observed affinities between populations that breed in Puka and Kukes or Dibra and Mat, but affinities between populations that breed in Erseka and Kruja, in Permet and Lushnja or in Vlora and Lezha, that are geographically far apart does not support this hypothesis. As noted above, these groups do not match with the geographical proximity between regions, so the phenomenon of isolation in distance does not observe, more is likely to be true hypothesis that between apiaries which are located in different regions, during the years have occurred exchange-migrations and genes immigrations, which have made that differences between them to be small. Wasn't expected, great differentiation of local bee populations in Albania, just as in study conducted about 20 years ago, this for reason of genes migrations and ways of collective movements of bees. With advanced technological methods we thought to realize differentiation of bees after 20 years but as it looks its homogeneity is maintained.

morphological traits. There is a kind of homogenization of bee populations, because of massive migration between numerous economies. References 1.

Ruttner F, Tassencourt L, Louveveaux J: Biometrical-Statistical analysis of the the geographic variability of Apis mellifera L. Apidologie. 1978, 9: 363-381.

2.

El-Aw A. M, Draz Kh. A, Eid Kh. S. A, Abo-Shara H: Measuring the morphological characters of honey bee (Apis mellifera L.) using a simple semi –automatic technique. Journal of American science. 2012, 8: 558-564.

3.

Ruttner F. Biogeography and Taxonomy of Honey bees. 1988.

4.

Nuri Q. Veçoritë morfologjike të bletës së vëndit. Buletini i Shkencave Bujqësore. 1963, 2: 126-150.

5.

Thomo K, Shehu L, Bajrami Z, Koni M: Diferencimi lokal i populacioneve të bletëve (A.m carnica) në Shqipëri. Buletini i Shkencave Bujqesore. 2002, 2: 75-82.

6.

Dedej S. Morphometric and alloenzymatic characterization in the Albania honey bee population (Apis mellifera L.). Apidologie. 1996, 27: 121-131.

7.

Johnson B. R, Nieh J. C: Modeling the adaptive role of negative signaling in honey bee intraspecific competition. Journal of Insect Behavior. 2010, 23: 459-471.

8.

Alpatov W. W. Biometrical studies on variation and races of the honey bee (Apis mellifera L.). Quarterly Review of Biology. 1929, 4: 1-58.

9.

Dupraw E. J. Non-Linnean taxonomy and the systematics of honey bees. Systematic Zoology. 1965, 14: 1-24.

Conclusion From obtained results, reinforced by respective analysis and interpretations, we conclude that local differentiation in Albanian bee population is not great. This differentiation cannot be explained by the phenomenon of isolation in distance. Human activities made during the last fifteen years like introduction of genes from neighboring countries into different geographic areas, or randomly exchanges of genes flow between apiaries that breed in different regions of Albania brought significant changes in variations of

10. Ruttner F. Geographical variability and classification. Bee Genetics and Breeding Academic Press. 1986, FL: 23-56. 11. Alpatov W W. The races of honey bees and their use in agriculture. Sredi Prirodi. 1948, 4: 1-65.

425