Food Chemistry 141 (2013) 3386–3392

Contents lists available at SciVerse ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Physicochemical properties of the Harenna forest honey, Bale, Ethiopia

Abera Belay a, W.K. Solomon b, Geremew Bultossa c, Nuru Adgaba d, Samuel Melaku e,⇑
a
Center for Food Science and Nutrition Program, Addis Ababa University, PO Box 1176, Addis Ababa, Ethiopia
b
University of Swaziland, Faculty of Agriculture, Department of Consumer Sciences, Luyengo M205, Swaziland
c
Department of Food Science and Technology, College of Agriculture, Private Bag 0027, Content Farm, Sebele, Gaborone, Botswana
d
Bee Research Unit, College of Food Science and Agriculture, King Saud University, PO Box – 1460, Riyadh 11451, Saudi Arabia
e
Department of Chemistry, Columbus State University, 4225 University Avenue, Columbus, GA 31907, USA

a r t i c l e i n f o a b s t r a c t

Article history: In this study, the physicochemical properties of the Harenna forest honey were characterised. The
Received 6 March 2013 Harenna forest honey moisture, reducing sugar, sucrose, water insoluble solids, ash, free acid, pH, HMF
Received in revised form 8 June 2013 contents, electrical conductivity and specific rotation were found to be 17.89 ± 1.02 g/100 g,
Accepted 10 June 2013
69.48 ± 1.72 g/100 g, 2.43 ± 1.02 g/100 g, 0.12 ± 0.08 g/100 g, 0.19 ± 0.09 g/100 g, 34.57 ± 4.80 meq/kg,
Available online 18 June 2013
3.87 ± 0.16, 0.84 ± 0.46 mg/1000 g, 0.70 ± 0.04 mS/cm and 132 ± 15.27 ½a20 D , respectively. All quality
indicators of honey from traditional and frame hives were within the criteria set by Codex Alimentarus
Keywords:
(CA), European Union (EU) and Ethiopian standard, except for water insoluble solids. The type of hives
Physicochemical properties
Honey
significantly affected the moisture (p < 0.01), reducing sugar (p < 0.05), ash (p < 0.05) and HMF
Harenna forest (p < 0.05) contents of the Harenna forest honey. The sampling location also significantly affected the
Ethiopia moisture (p < 0.001), water insoluble solids (p < 0.01), ash (p < 0.01), electrical conductivity (p < 0.001)
and specific rotation (p < 0.001) values of the Harenna forest honey. Significant correlations were
observed between moisture content and electrical conductivity (r = 0.76, p < 0.01), and electrical conduc-
tivity and specific rotation (r = 0.74, p < 0.01). Traditional hive has no negative effect on quality factors of
honey if honey harvesting, handling and processing is properly carried out.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Analysis of the physicochemical properties of honey is used to

Honey is the natural sweet substance produced by honeybees
from the nectar of blossoms or from the secretion of living parts
of plants or excretion of plant sucking insects on the living parts
of plants, which honeybees collect, transform and combine with
specific substances of their own, store and leave in the honeycomb
to ripen and mature (Codex Standard for Honey, 2001).
Physicochemical (moisture, reducing sugar, sucrose, water
insoluble, ash, free acid, hydroxymethylfurfural contents, pH, elec-
trical conductivity and specific rotation), sensorial and microbio-
logical characteristics are used to determine the quality of honey.
The physicochemical properties for a given honey is influenced
by the nectar types that the honey bee used, geographical ecology
(climatic and soil) and postharvest honey handling practices.
⇑ Corresponding author. Tel.: +1 7065078476; fax: +1 7065693133.
E-mail addresses: ab.berabelay@gmail.com (A. Belay), solowkj@yahoo.com
(W.K. Solomon), bultosageremew@yahoo.com (G. Bultossa), nuruadgaba@gmail.
com (N. Adgaba), abegaz_samuel@columbusstate.edu, samuelmelaku@yahoo.com
(S. Melaku).
verify the authenticity of the product and to reveal the possible
presence of artificial components or adulterants (Cantarelli,
Pellerano, Marchevsky, and Camiña (2008), Elke Anklam (1998)).
Honey authenticity is defined by Codex Alimentarius, EU and a
few national legislations.
Physicochemical analysis of honey is also important in the glo-
bal trade since honey is used in an ever-increasing array of food
products. The driving factors for wide use of honey as food ingre-
dient is due to nutritional value and non-nutrient functionalities
(sweetness, humectancy, viscosity, flavour enhancement, colour,
hygroscopicity, miscibility and spreadability) (Bogdanov, 2001).
About 42,000 tones of honey are estimated to be produced in
Ethiopia annually that makes the country first in Africa and one
of the ten major honey producing countries in the world (Nuru,
1999).
Although honey is widely produced and consumed in Ethiopia,
there is no information reported on the physicochemical properties
of honey harvested from the Harenna forest, Bale, Ethiopian. This
study was conducted with the aim of evaluating the physicochem-
ical properties of honey harvested from traditional and frame hives

0308-8146/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2013.06.035
 A. Belay et al. / Food Chemistry 141 (2013) 3386–3392
located in the Harenna forest compared to international standards
available in this field.
2. Materials and methods
Sixteen honey samples (four traditional and four frame hives
from two study areas) were collected from the Chiri and Waberro
sampling locations in the Harenna forest, Bale, Ethiopia, during the
major honey flow season; between the 18th of January, 2010 and
the 20th of February, 2010; using randomized lottery sampling
methods.
2.1. Moisture content
Moisture content (g/100 g) was determined using a refractome-
ter (Leica Mark II plus Abbe’ refractometer, USA) at 20 °C using
refractive index of distilled water (1.3330) as a reference. The
refractive index reading was converted to moisture content
(g/100 g) using Table 969.38 (AOAC, 1990 method number 969.38).
2.2. Apparent reducing sugar content
Apparent reducing sugar content was determined by the mod-
ified Lane–Eynon procedure involving the reduction of Soxhlet’s
modification of Fehling’s solution by titration at boiling point
against a solution of reducing sugars in honey solution (dissolving
about 2 g homogeneous honey sample into 100 mL distilled water
from which 50 mL was diluted to 100 mL) using methylene blue as
an internal indicator of Harmonised IHC method (Bogdanov, 2002):
2  1000
C ðg=100 gÞ ¼
W2  Y2
where C = g invert sugar per 100 g honey, W2 = weight (g) of honey
sample and Y2 = volume (mL) of diluted honey solution consumed
in the titration.
2.3. Apparent sucrose content
Apparent sucrose content (g sucrose/100 g honey) was deter-
mined after inversion of honey solution (Section 2.2) with 6.34 M
HCl as described in Harmonised IHC (Bogdanov, 2002). Apparent
sucrose (g/100 g) = (A  B)  0.95. Where A = invert sugar content
after inversion and B = invert sugar content before inversion both
in g/100 g honey.
2.4. Water insoluble solids content
Water insoluble solids content were determined as residue
mass (m) obtained by filtering honey solution (m2) (about 20 g
honey/200 mL distilled water) through sintered glass (PYREX, UK)
porosity No. 2 (15–40 microns) and drying the residue at 135 °C
for 1 h (Bogdanov, 2002):
m
Insoluble solidsðg=100 gÞ ¼  100
m2
where m = mass of dried insoluble matter, m2 = mass of honey
taken.
2.5. Ash content
Ash content was determined by taking about 10 g honey sam-
ple, after addition olive oil (2 drops), evaporation (300 °C) on heat-
ing plate (Stuart, UK), carbonisation in a blue Bunsen burner
(400 °C) and ashing in a muffle furnace (Therm Concept LAC,
Germany) at 600 °C until ashing complete (AOAC, 1990 method
920.181):
3387
ðm3  m1 Þ
Ash ðg=100 gÞ ¼  100
m2
where m1 = weight of ashing dish, m2 = weight of honey taken and
m3 = weight of ashing dish + ash.
2.6. Free acidity and pH
Free acid (meq of acid/1000 g) was determined by dissolving
honey sample (10 g/75 mL distilled water) and titrating with
standardised 0.1 M NaOH to pH 8.3 using pH glass electrode at-
tached to pH metre (Consort C861, Belgium) as end point indicator.
pH was determined by using glass electrode after calibration with
standard buffer solution pH 4, 7 and 10 (AOAC, 1990 method num-
ber 962.19).
2.7. Hydroxymethylfurfural (HMF)
HMF was determined from about 5 g honey sample (clarified
with Carrez I and Carrez II solutions) by subtracting the absorbance
measured at 284 nm for HMF in the honey sample solution against
the absorbance of reference (the same honey solution treated with
sodium bi-sulphite, 0.2%) at 336 nm using UV–Vis spectrophotom-
eter, Nicolet evolution 100 England (AOAC, 1990 method number
980.23):
HMFðmg=1000 ghoneyÞ ¼ ðA284  A336 Þ  149:7  5  D  W
Factor ¼ 149:7 ¼ ð126=16830Þð1000=10Þð1000=5Þ
where A284 = absorbance at 284 nm, A336 = absorbance at 336 nm,
126 = molecular weight of HMF, 16,830 = molar absorptivity of
HMF at 284 nm, 1000 = conversion of g into mg, 10 = conversion
of 5 into 50 mL, 1000 = conversion g honey into 1000 g (kg),
5 = nominal sample weight, D = Dilution factor and W = weight in
g of the honey sample.
2.8. Electrical conductivity
Electrical conductivity was determined by conductivity cell
(HAAKE, Germany) by dissolving about 20 g honey sample into
100 mL distilled water placing the sample beaker into thermostat
controlled water bath (Consort C861, Belgium) at (20 °C) and as de-
scribed in Harmonised IHC method (Bogdanov, 2002) and conduc-
tance was given in mS/cm as follows:
SH ¼ K  G
where SH = electrical conductivity of the honey solution in mS/cm,
K = cell constant in cm1 and G = conductance in mS.
2.9. Specific rotation
Specific rotation was determined by measuring the angular
rotation of a clear (clarified by Carrez I and Carrez II solutions), fil-
tered aqueous honey solution (10 g dissolved into 100 mL solution)
by 2-dm polarimeter (KRUSS, Germany) at 20 °C as described in
Harmonised IHC method (Bogdanov, 2002) and the specific rota-
tion was calculated:
a  100
½a20
D ¼
Lg
where a = angular rotation found, L = length in decimeters of
polarimeter tube and g = grams of dry matter sample mass taken
and D = sodium D line.
2.10. Statistical Analysis: The data was analysed by ANOVA
(SAS, 2002). Means difference was compared using the Tukey-
Karmer test (p < 0.05 and 0.01). Correlation analysis (p < 0.05 and
3388 A. Belay et al. / Food Chemistry 141 (2013) 3386–3392

Table 1
Physicochemical properties of the Harenna forest honey, the effect of locations (Chiri and Wabero) and hive type (traditional and frame hives) (Values are mean ± sd) (n = 16).

Variables Units Hive type Location

p value Traditional Frame p value Chiri Wabero
Moisture g/100 g ⁄⁄
18.28 ± 1.08a 17.50 ± 0.80b ⁄⁄⁄
17.29 ± 0.87a 18.49 ± 0.78b
Reducing sugar g/100 g ⁄
70.14 ± 2.21a 68.82 ± 0.53b ns 69.75 ± 2.24 69.20 ± 0.96
Sucrose g/100 g ns 2.19 ± 0.45 2.66 ± 1.36 ns 2.56 ± 1.19 2.30 ± 0.85
Water insoluble g/100 g ns 0.14 ± 0.09 0.09 ± 0.06 ⁄⁄
0.16 ± 0.10a 0.08 ± 0.01b
Ash g/100 g ⁄
0.20 ± 0.10a 0.17 ± 0.06b ⁄⁄
0.17 ± 0.07a 0.21 ± 0.10b
Free acid meq/1000 g ns 35.80 ± 5.60 33.33 ± 3.60 ns 35.18 ± 6.37 33.96 ± 2.48
pH pH units ns 3.87 ± 0.08 3.87 ± 0.22 ns 3.86 ± 0.09 3.88 ± 0.22
HMF mg/1000 g ⁄
1.01 ± 0.47a 0.68 ± 0.40b ns 0.92 ± 0.54 0.76 ± 0.38
Electrical conductivity mS cm1 ⁄⁄
0.71 ± 0.03a 0.68 ± 0.04b ⁄⁄⁄
0.67 ± 0.02a 0.73 ± 0.03b
Specific rotation ½a20
⁄⁄⁄
123.12 ± 2.50a 141.25 ± 16.88b ⁄⁄⁄
141.25 ± 16.88a 123.12 ± 2.50b
D

HMF = hydroxymethylfurfural; n = number of samples; ns = not significant; sd = standard deviation ⁄ significant at p < 0.05; ⁄⁄ significant at p < 0.01; ⁄⁄⁄
significant at
p < 0.001. Means in a row, for traditional and frame hive; and for Chiri and Wabero, with different letters are significantly different.

0.01) was conducted to determine the associations between honey
quality parameters.
3. Results and discussion
Table 1 shows the results of the physicochemical analysis of the
Harenna forest honeys compared with international honey quality
standards.
3.1. Moisture
The mean moisture content of the Harenna forest honey was
17.89 ± 1.02 g/100 g (15.87–19.35 g/100 g), Table 1. Moisture con-
tent of honey samples from traditional and frame hives ranged
from 15.87 to 19.35 g/100 g (mean 18.28 ± 1.08 g/100 g) and
16.63 to 19.19 g/100 g (mean 17.50 ± 0.80 g/100 g), respectively.
Moisture content of honey from Chiri and Wabero locations ranged
from 15.87 to 18.81 g/100 g (mean 17.29 ± 0.87 g/100 g) and 17.05
to 19.35 g/100 g (mean 18.49 ± 0.78 g/100 g), respectively, Table 1.
A significant difference in moisture content was observed among
honey samples due to hive types (p < 0.01) and locations
(p < 0.001).
This investigation was in line with the findings of Finola, Lasag-
no, and Marioli (2007), Rehman, Khan, and Maqbool (2008) and
Cantarelli et al. (2008) that water content is strictly related to de-
gree of maturity in the hive, botanical origin of the honey sample,
harvesting techniques and extraction from the comb in relation to
the ripening process by bees. Honey samples of the study area
were dominated and enriched by different honey plants: Syzgium
guineense, Guania longispicata, Trifolium spp., Hypoythes spp., Croton
spp., Ekbergia capnsis, Guizotia spp., and Unknown spp. Separation
honey from the combs was performed using honey extractor for
frame hives and honey strainer for traditional hives.
Muli, Munguti, and Raina (2007) reported moisture content of
honey ranging from 16.00 to 21.20 g/100 g, for the Kenyan honeys.
Compared to Kenyan honey, the Harenna forest honey contained
lower moisture, 17.89 ± 1.02 g/100 g. The mean moisture content
of honey, in this study, was also lower than the moisture content
of the country’s average (20.6 g/100 g), Nuru (1999). The maxi-
mum limit for moisture content set by the International Honey
Commission is 20 g/100 g (IHC, 2002).
In routine honey control carried out by the Institute for Honey
Analysis (IHA) during the years 1989–97 on 30,000 honey samples,
91–95% of honeys distributed had less than 20 g/100 g. From the
European countries the standard for Switzerland is less than
20 g/100 g (Bogdanov, Lullmann, Martin, Ohe, & Russmann,
1999). Many national beekeeping organisations of Europe:
Germany, Belgium, Austria, Italy and Spain set out moisture
content of maximum values of 17.50–18.50 g/100 g for special
classes of quality honey (Bogdanov et al., 1999). Thus, the moisture
content of Harenna forest honey (17.89 ± 1.02 g/100 g) satisfies the
special class honey of beekeeping organisation of Europe.
According to the Ethiopian Standard, the moisture content of
honey grouped into three grades: ‘Grade A’, 17.50–19.00 g/100 g;
‘Grade B’, 19.10–20.00 g/100 g; and ‘Grade C’, 20.10–21.00 g/
100 g. The Harenna forest honey can be grouped as ‘Grade A’ honey
based on the Ethiopian Standard (2005).
The CA and EU demand a maximum of 21 g/100 g of honey
(Bogdanov, 2001). Even though there is a statistically significant
difference in hive types and locations; all the honey samples col-
lected from the Harenna forest satisfied the CA, EU and Ethiopian
standards with respect to moisture content.
3.2. Reducing sugar
The mean reducing sugar content of Harenna forest honey is
69.48 ± 1.72 g/100 g (Table 1). The mean value of reducing sugar
content by hive types and locations are presented in Table 1.
Reducing sugar content of honey the traditional and frame hives
ranged from 67.79 to 74.62 g/100 g (mean 70.14 ± 2.21 g/100 g)
and 67.94 to 70.00 g/100 g (mean 68.82 ± 0.53 g/100 g), respec-
tively. The mean reducing sugar content of honey samples col-
lected from Chiri and Wabero were 69.75 ± 2.24 g/100 g (67.94–
74.62 g/100 g) and 69.20 ± 0.96 g/100 g (67.79–70.50 g/100 g),
respectively. The mean reducing sugar content of traditional hive
honey was significantly (p < 0.05) higher in value than frame hive.
All traditional hives of the Harenna forest were hanging on selected
tallest trees, 30–40 m far from the ground; while the frame hives
were on the ground. There is a general truth that: as the height in-
crease from the ground, the possibility of evaporation increase.
Thus the position of traditional hives helps to concentrate the solid
components of the honey which is mainly resulted in an increment
in reducing sugar content. Moreover, the variability in mean reduc-
ing sugar content between traditional and frame hives could be
due to the difference in the sugars present in the nectar and en-
zymes present in the bee and nectar. However, there was no signif-
icant difference (p > 0.05) due to locations.
The Harenna forest honey showed higher values in mean reduc-
ing sugar content than the national mean value, 65 g/100 g, Nuru
(1999). Cantarelli et al. (2008) reported a mean reducing sugar
content of 68.80 g/100 g for Argentinean honeys.
According to Bogdanov (2009) sugars are the main constituents
of honey, comprising about 95% of honey dry weight. This study
also showed that reducing sugars were the main constituents of
the dry weight of honey. This is in harmony with the findings of
Lazaridou, Biliaderis, Bacandritsos, and Sabatini (2004) who
 A. Belay et al. / Food Chemistry 141 (2013) 3386–3392
showed that honey is a concentrated solution of monosaccharides,
in which 60–85% of the product is composed of both fructose and
glucose.
The Harenna forest honey can be defined as a blossom honey
with higher concentration of fructose than glucose. This was
proved by analysis result of reducing sugar and specific rotation
(laevo-rotatory). Codex Alementarious categorised honeydew hon-
ey or blends of honeydew honey and blossom honey to consist
about 45 g/100 g reducing sugar (Bogdanov, 2001).
The reducing sugar result of this study satisfied the CA, EU and
the Ethiopian standard. The CA, EU and the Ethiopian standard
have minimum limit of 65 g/100 g reducing sugar in honey (Bogda-
nov, 2001; Ethiopian Standard, 2005).
3.3. Sucrose
Analysis of sucrose content is used to detect the adulteration of
honey with table sugar or to determine the amount of sucrose nat-
urally found in a given honey sample. The mean sucrose content of
honey sample from the study area was 2.43 ± 1.02 g/100 g
(Table 1).
The mean sucrose content of honey samples from the tradi-
tional and frame hives were 2.19 ± 0.45 g/100 g (1.70–2.82 g/
100 g) and 2.66 ± 1.36 g/100 g (0.75–4.52 g/100 g), respectively,
Table 1. Honey samples from Chiri showed a sucrose content rang-
ing from 0.75 to 4.52 g/100 g with a mean value of 2.56 ± 1.19 g/
100 g, while honey samples from Wabero ranged from 1.16 to
4.04 g/100 g (mean 2.30 ± 0.85 g/100 g). Results of this study
showed that there was no significant difference (p > 0.05) between
means of sucrose content of honey samples across hive types and
locations.
Cantarelli et al. (2008) and Muli et al. (2007) reported sucrose
content of 4.05 g/100 g and 0.9–2.2 g/100 g, for the Argentinean
and Kenyan honeys, respectively. The results of their studies
showed that sucrose content of honey samples from Argentina
and Kenya were within the limits of the quality criteria, 5 g/100 g.
The mean value of sucrose in the Harenna forest honey is lower
than the maximum limits set by IHC (2002) and the national aver-
age, 3.6 g/100 g, Nuru (1999). The result of the chemical analyses of
sucrose revealed that all the honey samples collected from the Har-
enna forest satisfied the CA, EU and the Ethiopian standards. The
CA, EU and Ethiopia standard have a maximum limit of 5 g/100 g
of honey (Bogdanov, 2001; Ethiopian Standard, 2005).
3.4. Water insoluble solids
The water insoluble component of honey includes wax, pollen,
honey-comb debris, bees and filth particles. The measurement of
water insoluble solids is important to detect honey impurities that
are higher than the permitted maxima.
The mean water insoluble content of the Harenna forest honey
samples was 0.12 ± 0.08 g/100 g (0.05–0.38 g/100 g), Table 1. The
mean water insoluble solid of the Chiri and Wabero honey samples
were 0.16 ± 0.10 g/100 g (0.05–0.38 g/100 g) and 0.08 ± 0.01 g/
100 g (0.05–0.11 g/100 g), respectively, Table 1. The water insolu-
ble solids content showed a significant difference (p < 0.01) be-
tween the two locations. The result of water insoluble solids of
honey samples from frame and traditional hives ranged from
0.05 to 0.26 g/100 g with a mean value of 0.09 ± 0.06 g/100 g and
0.07 to 0.38 g/100 g with a mean 0.14 ± 0.09 g/100 g; respectively.
There was no significant difference (p > 0.05), in water insoluble
content, between the two hive types.
The mean value (0.12 g/100 g) of water insoluble component of
the Harenna forest honey was above the limit of CA, EU and Ethi-
opian standards. CA, EU and Ethiopian standard has a maximum
limit of 0.1 g/100 g for extracted honey (Bogdanov, 2001; Ethiopian
3389
standard, 2005) and 0.5 g/100 g for pressed honey as reported by
CA (Bogdanov, 2001).
The higher water insoluble solid content might be attributed to
the knowledge of beekeepers and lack of access to quality extrac-
tion and straining equipment. Giving a practical training to local
beekeepers about the way to harvest, handle, process, package
and market honey and facilitating access to proper processing
equipment will help to solve the problem.
3.5. Ash
The mean ash content of the Harenna forest honey was
0.19 ± 0.09 g/100 g (0.05–0.35 g/100 g), Table 1. The average ash
content of honey samples for the traditional and frame hives were
0.20 ± 0.10 g/100 g (0.05–0.37 g/100 g) and 0.17 ± 0.06 g/100 g
(0.08–0.26 g/100 g), respectively, Table 1. The mean ash content
of honey samples from Chiri and Wabero sites were
0.17 ± 0.07 g/100 g (0.05–0.26 g/100 g) and 0.21 ± 0.10 g/100 g
(0.08–0.37 g/100 g), respectively. The ash content of honey sam-
ples of traditional hives was significantly (p < 0.05) higher than
that of frame hives, and the values for Wabero was higher than that
of Chiri (p < 0.01).
The difference in ash content across location was expected to
happen due to the variability of soil type, concentration of minerals
found in the nectar on different apiaries and disparate nature of the
botanical origin of honey samples of Chiri and Wabero. This was in
agreement with the findings of Bogdanov (2007) that environmen-
tal, geographical and botanical factors influence the trace element
content of honey. The difference in ash content of honey samples
between hive types could be due to the honey harvesting practices
used to separate honey from the combs and the type of honey
plants gathered by honeybees. This was in line with the work of
Finola et al. (2007) that the variability in ash content of honeys
could be due to harvesting, beekeeping techniques and the mate-
rial collected by the bees during foraging on flora.
The mean values of ash reported for honeys from Argentina and
Nigeria were 0.11 g/100 g (Cantarelli et al., 2008) and 0.28 ± 0.03 g/
100 g (Omafuvbe & Akanbi, 2009), respectively. These values were
higher than the ash content of the Harenna forest honeys. The
mean ash content of honey produced in the study area was also
lower than the national average value, 0.23 g/100 g, Nuru (1999).
The maximum limit for mineral content in honey is 0.6 g/100 g
(Bogdanov, 2001; Ethiopian standard, 2005). The results of this
study showed that the mean ash content of the Harenna forest
honey was quite lower and found to be within the accepted range
of the national, CA and EU standards. The lower content of ash from
Harenna forest is an indication for pollution free environment. This
was in line with Anklam (1998) that ash content in honey samples
could give an indication of environmental pollution.
3.6. Free acid
Acidity of honey is due to the presence of organic acids and
inorganic ions such as the gluconic acid with their lactones or es-
ters, phosphate and chloride. Acid measurement is useful for eval-
uation of honey fermentation, authentication of unifloral honeys
and differentiating nectar from honeydew honeys (Nandaa,
Sarkara, Sharmaa, & Bawab, 2003).
The mean free acid content of the Harenna forest honey
samples was 34.57 ± 4.80 meq/kg (25.49–48.81 meq/kg), Table 1.
The average acid content of honey samples from the traditional
and frame hives were 35.80 ± 5.60 meq/kg (30.79–48.81 meq/kg)
and 33.33 ± 3.60 meq/kg (25.49–39.21 meq/kg), respectively,
Table 1. The mean acid content of honey sample from Chiri and
Wabero sites were 35.18 ± 6.37 meq/kg (25.49–48.81 meq/kg)
and 33.96 ± 2.48 meq/kg (30.79–39.21 meq/kg), respectively.
3390 A. Belay et al. / Food Chemistry 141 (2013) 3386–3392
There was no significant difference in free acid content of honey
samples due to hive types and sampling locations (p > 0.05). The
mean free acid content of the Harenna forest honey was compara-
ble to the Venezuelan honey, 35.14 meq/kg (24.40–54.55 meq/kg),
Vit P. et al. (2009).
According to Baltac and Candan (2007), lower value of acid indi-
cates absence of undesirable fermentations. The mean free acid va-
lue of the current study was below the national average, 39.9 meq/
kg (Nuru, 1999), and satisfied the CA, EU and Ethiopian standards.
The EU and Ethiopian standards have more strict norms compared
to CA standard. The maximum limit for free acid set by the CA is
50 meq/kg of honey, while the EU and Ethiopian standard is
40 meq/kg of honey.
3.7. pH
In general, honey is acidic in nature irrespective of its variable
geographical origin. The average pH value of the Harenna Forest
honey samples was 3.87 ± 0.16 (3.45–4.18), Table 1. The Harenna
forest honey was found to be acidic in character. The mean pH
value of honey sample from traditional hives was 3.87 ± 0.08
(3.74–3.99), while honey from frame hive was 3.87 ± 0.22
(3.45–4.18), Table 1. The pH value of the Chiri and Wabero honey
samples ranged from 3.73 to 3.98 with a mean value of 3.86 ± 0.09
and 3.45 to 4.18 with a mean of 3.88 ± 0.22, respectively (Table 1).
There was no statistically significant difference (p > 0.05) in pH
among honey samples due to hive types or locations.
The pH value of honey samples from Artisanal (Madrid) and
Cordoba (Argentina) were reported to be in the range of 3.63–
5.01 (Soria, Gonzalez, Lorenzo, Martınez-Castro, & Sanz, 2004)
and 3.14–5.05 (Baroni et al., 2009), respectively. According to Che-
frour et al. (2009), honeys with pH range from 3.5 to 4.5 are con-
sidered to be a blossom honey, while honey samples with a pH
above 5 to be of low quality. Thus, the Harenna forest honey can
be categorised as blossom honey and lies within the reported qual-
ity pH range, 3.5–4.5 (Chefrour et al., 2009).
3.8. Hydroxymethylfurfural (HMF)
The shelf-life of honey is established worldwide according to
HMF content criteria set by national regulations. Packaging plants
are obliged to comply with the requirement when establishing the
use by date to be printed on their labels. Compliance with HMF
content restrictions is important as it provides consumers with a
guarantee that the honey they are buying has only been altered
minimally (Escriche, Visquert, Carot, Domenech, & Fito, 2008).
Fresh honey does not contain HMF (Bogdanov, Ruoff, & Oddo,
2004); or contains only traces of HMF, which is an important crite-
rion for the evaluation of storage time and heat damage (Ruoff &
Bogdanov, 2004).
The HMF value of the Harenna forest honey samples ranged
from 0.00 to 1.71 mg/kg with a mean value of 0.84 ± 0.46 mg/kg
(Table 1). The mean value of HMF in the traditional and frame
hives were 1.01 ± 0.47 mg/kg (0.39–1.71 mg/kg) and
0.68 ± 0.40 mg/kg (0.00–1.41 mg/kg), respectively. The average
content of HMF in the Chiri and Wabero honey samples were
0.92 ± 0.54 mg/kg (0.00–1.71 mg/kg) and 0.76 ± 0.38 mg/kg
(0.39–1.41 mg/kg), respectively. There was no significant differ-
ence (p > 0.05) in HMF content of honey samples due to locations.
However, there was a significant difference (p < 0.05) in HMF con-
tent of honey samples due to hive types.
Muli et al. (2007), Soria et al. (2004); and Vit et al. (2009) re-
ported HMF content ranging from 3.70 to 389.36 mg/kg, 0.00 to
15.65 mg/kg and 17.70 to 631.73 mg/kg honey with a mean value
of 162.71 mg/kg for honey samples collected from Kenya, Madrid
and Venezuelan, respectively. The mean HMF value of the Harenna
Forest was by far lower than the values reported for Kenya, Madrid,
Venezuelan, Algeria and the national average, 32.4 mg/kg (Nuru,
1999).
Even though the honey samples have a significant difference
across hive types, all the honey samples collected from the Haren-
na Forest were far below the maximum acceptable level of CA, EU,
Ethiopian standards, the international trade and the sever limits of
the European bee federation.
3.9. Electrical conductivity
Electrical conductivity depends on ash, organic acids, proteins,
some complex sugars and polyols contents, and varies with botan-
ical origin. Electrical conductivity is widely used for discrimination
between honeydew and blossom honeys and also for the charac-
terisation of unifloral honeys (Chefrour et al., 2009).
The electrical conductivity value of Harenna forest honey
ranged from 0.63 to 0.79 mS/cm with a mean value of 0.70 ± 0.04
(Table 1). The average values of electrical conductivity for tradi-
tional and frame hives were 0.71 ± 0.03 mS/cm (0.69–0.78 mS/
cm) and 0.68 ± 0.04 mS/cm (0.63–0.79 mS/cm), respectively. The
mean electrical conductivity values of honey samples collected
from Chiri and Waberro locations were 0.67 ± 0.02 mS/cm
(0.63–0.71 mS/cm) and 0.73 ± 0.03 mS/cm (0.69–0.79 mS/cm),
respectively.
Electrical conductivity varied significantly due to hive types
(p < 0.01) and locations (p < 0.001). The mean electrical conductiv-
ity of traditional hive honey was higher than frame hive honey and
the mean electrical conductivity of Waberro was higher than Chiri
honey.
Reports from Codex Alimentarius Committee on Sugars (2001)
and European Commission (2002) stated that the botanical origin
of honeys is often classified into two classes: blossom and honey-
dew honey. Categorising the samples to blossom or honeydew ori-
gin was based on the measurement of the electrical conductivity.
Honeys with electrical conductivity values higher than 0.8 mS/cm
are considered as honeydew honeys, while those with lower values
than 0.8 mS/cm are blossom honeys or blends of blossom with
honeydew honey (Bogdanov, 2007). Accordingly, the honey sam-
ples in this study are categorised as blossom honeys. Honey sam-
ples with an electrical conductivity value ranging from 0 to
0.8 mS/cm are classified as honeys produced from nectar without
mixture compounds of honeydews and nectar blossom. The honey-
dew honeys are characterised by their very dark colour and high
values of pH, ash content and electrical conductivity.
According to Soria et al. (2004), the electrical conductivity of
honeys from Madrid was ranging from 0.119 to 1.515 mS/cm.
The electrical conductivity of honey samples in the Harenna forest
honey was lower than the results reported by Soria et al. (2004).
The international standards for electrical conductivity set a
maximum limit of 0.8 mS/cm for most nectar honeys with several
exceptions, among which is Eucalyptus honey. Even though there
is a significant difference in electrical conductivity due to hive
types and locations, the electrical conductivity of the Harenna for-
est honey samples satisfied the CA and EU standards.
3.10. Specific rotation
The determination of the specific rotation by means of a polar-
imeter is useful for the differentiation between honeydew (dextro-
rotatory/positive values) and blossom honeys (laevo-rotatory/neg-
ative values) (Oddo & Piro, 2004). This may also help to detect hon-
ey adulteration. According to Bogdanov et al. (2004) natural honey
has the property of rotating the plane of polarised light to the left.
Some sugars (fructose) exhibit a negative optical rotation, laevoro-
tatory, while others (glucose) a positive one, dextro-rotary. The
 A. Belay et al. / Food Chemistry 141 (2013) 3386–3392 3391

Table 2
Pearson correlation coefficients among the analysed parameters (moisture, reducing sugar, sucrose, WIS, ash, free acid, pH, HMF, EC and SR.).

Parameters RS Sucrose WIS Ash Acid pH HMF EC SR

Moisture .397 .132 .537 .356 .240 .155 .342 .762(⁄) .559
RS .303 .203 .324 .570 .135 .060 .029 .347
Sucrose .412 .080 .097 .251 .285 .230 .272
WIS .168 .065 .043 .506 .472 .557
Ash .351 .280 .417 .246 .180
Acid .367 .371 .228 .327
pH .092 .397 .021
HMF .318 .359
EC .742(⁄)

EC = electrical conductivity; HMF = hydroxymethylfurfural; r = correlation; RS = reducing sugar; SR = specific rotation; WIS = water insoluble solids. ⁄ correlation is significant
at 0.01 level.

overall specific rotation of honey depends on the concentration of
the various sugars present in the honey.
Specific rotation (½a20D ) of the Harenna forest honey samples
ranged from (160) to (120), with a mean value of
132 ± 15.02 (Table 1). The mean specific rotation values of the
traditional and frame hives were 123.12 ± 2.50 (125 to 120)
and 141.25 ± 16.88 (160 to 125), respectively, (Table 1). The
average specific rotation value of honey samples from Chiri and
Wabero were 141.25 ± 16.88 (160 to 125) and
123.12 ± 2.50 (125 to 120), respectively, (Table 1). Statistically
significant differences (p < 0.001) were observed in specific rota-
tions of honey samples due to hive types and locations.
The results of this study indicated that honey samples from the
Harenna forest were from blossom honey, rotating negative (laevo-
rotatory) (Table 1). Mean value of specific rotation of honey from
frame hives was more laevo-rotatory than those from traditional
hives. Honey samples from Chiri were more laevo-rotatory than
Wabero. This feature makes the honey to have better sweetness
and to stay liquid for reasonable amount of time than honey with
a higher concentration of glucose. This property protects rapid
granulation of the Harenna forest honey.
3.11. Correlation between the physicochemical properties of honey
The Pearson correlation coefficients between all parameters are
presented in Table 2. There were significant correlations between
moisture content and electrical conductivity r = 0.76 (p < 0.01).
Electrical conductivity and specific rotation were also significantly
correlated, r = 0.74 (p < 0.01). The relationship between electrical
conductivity and specific rotation is one of the most important
relations for honey classification and sorting into the individual
honey groups. The high correlation coefficient (r = 0.74) value indi-
cated that exact classification of honey must be carried out not
only by measuring the conductivity but also in relation to optical
rotation of individual honey groups. Other parameters had shown
numerical strong and weak correlation between variables.
The strong significant correlation between electrical conductiv-
ity and moisture could be due to the dependable nature of electri-
cal conductivity on honey moisture. This was supported by
Bogdanov (2001) that upon the addition of water, honey showed
remarkable changes in all of its physical properties, with major
changes occurring in conductance. Pure honey is characterised by
a low conductance value. If honey is adulterated with water or sat-
urated sugar solutions, it will display greater conductance than
pure honey.
The current study correlation finding between water and elec-
trical conductivity was also in agreement with the work of Guler,
Bakan, Nisbet, and Yavuz (2007) that electrical conductivity
(r = 0.551) were significantly correlated with water. The significant
correlation between electrical conductivity and specific rotation
(r = 0.74) (p < 0.01) of Harenna forest honey samples was also
supported by the works of Pridal and Vorlova (2002) from Czech
Republic honey that there was a significant correlation between
electrical conductivity and specific rotation (r = 0.86).
4. Conclusion
The Harenna forest honey moisture, reducing sugar, sucrose,
ash, free acid, pH, HMF contents, electrical conductivity and spe-
cific rotation values satisfied the CA, EU and Ethiopian standards;
except for the water insoluble solids. Thus, traditional honey
harvesting and processing methods do not seem to have negative
effects on quality factors of honey.
The synergetic effect of location and hive types showed signifi-
cant differences in the values of reducing sugar, ash, acidity, HMF
and specific rotation; whereas moisture, sucrose, water insoluble
solid contents, pH, and electrical conductivity did not.
The high correlation coefficient (r = 0.74) between electrical
conductivity and specific rotation indicated that exact classifica-
tion of honey must be carried out not only by measuring the con-
ductivity but also in relation to optical rotation of individual honey
groups.
On contrary to the common misconception that honey har-
vested and processed through traditional methods is generally of
low quality, the Harenna forest honey which is largely coming
from traditional hives satisfied the quality parameters set by the
CA, EU and Ethiopian standards.
Acknowledgements
Financial support from Adama Science and Technology Univer-
sity, and SOS SAHAL-Farm Africa is thankfully acknowledged. The
authors would also like to thank Prof. Dr. Herbert Eichele, Mr.
Ben Iriwn, Mr. Belay Asfaw, Mr. Lulu Likassa and Hirut Abebe with-
out whom this research was almost impossible.
References
Anklam, E. (1998). A review of the analytical methods to determine the
geographical and botanical origin of honey. Food Chemistry, 63, 549–562.
AOAC (Association of Official Analytical Chemists) (1990). Official methods of the
analysis of official analytical chemists. (15th ed.) Vol. II, USA: Virginia.
Baltac, C., & Candan, F. (2007). Biological activities and chemical composition of
three honeys of different types from Anatolia. Food Chemistry, 100, 526–534.
Baroni, M. V., Arrua, C., Nores, M. L., Faye, P., Diaz, M. P., Chiabrando, G. A., et al.
(2009). Composition of honey from Cordoba (Argentina): Assessment of North/
South provenance by chemometrics. Food Chemistry, 114, 727–733.
Bogdanov S. (2001) Codex Alimentarius: Draft Revised Standard for Honey Bee
Department. Federal Dairy Research Institute, 3003 Bern, Switzerland.
Bogdanov S. (2002) Harmonized Methods of the International Honey Commission.
Swiss Bee Research Center, CH-3003 Bern, Switzerland.
Bogdanov S. (2007) Authenticity of Honey and Other Bee Products: State of the Art.
Bulletin USAMV-CN, 63–64.
Bogdanov S. (2009) Honey composition. Bee Product Science. URL <http://www.bee-
hexagon.net>. Accessed 25.02.13.
3392 A. Belay et al. / Food Chemistry 141 (2013) 3386–3392
Bogdanov S., Lullmann C., Martin P., Ohe W. V., Russmann H. (1999) Honey Quality
and International Regulatory Standards: Review of the International Honey
commission. Swiss Bee Research Center. URL <http://www.db_alp.admin.ch/
en/publication/pubdetail.php?id=17750>. Accessed 24.02.13.
Bogdanov, S., Ruoff, K., & Oddo, L. P. (2004). Physico-chemical methods for the
characterisation of unifloral honeys: A review. Apidologie, 35, 4–17.
Cantarelli, M. A., Pellerano, R. G., Marchevsky, E. J., & Camiña, J. M. (2008). Quality of
honey from Argentina: Study of chemical composition and trace elements.
Journal of the Argentine Chemical Society, 96, 33–41.
Chefrour, A., Draiaia, R., Tahar, A., Ait, K. Y., Bennadja, S., & Battesti, M. J. (2009).
Physicochemical characteristics and pollen spectrum of some north-eastern
Algerian honeys. African Journal of Food, Agriculture, Nutrition and Development,
9, 1276–1293.
Codex Alimentarius Committee on Sugars. (2001) Codex standard 12, Revised Codex
Standard for Honey. Standards and Standard Methods, 11, 1-7.
Escriche, I., Visquert, M., Carot, J. M., Domenech, E., & Fito, P. (2008). Effect of honey
thermal conditions on hydroxymethylfurfural content prior to pasteurization.
Food Science and Technology International, 14, 29–35.
Ethiopian Standard. (2005) Honey Specification, Addis Ababa, Ethiopia.
European Commission. (2002) Council Directive 2001/110/EC of 20th, December
2001 relating to honey. Official Journal of the European Communities, L10, 47–52.
Finola, M. S., Lasagno, M. C., & Marioli, J. M. (2007). Microbiological and chemical
characterization of honeys from central Argentina. Food Chemistry, 100,
1649–1653.
Guler, A., Bakan, A., Nisbet, C., & Yavuz, O. (2007). Determination of important
biochemical properties of honey to discriminate pure and adulterated honey
with sucrose (Saccharumofficinarum L.) syrup. Food Chemistry, 105, 1119–1125.
IHC (2002) Harmonized Methods of the International Honey Commission. URL <http://
www.apis.admin.ch/english/host/pdf/honey/HMFabstract.pdf>). Accessed
21.02.13.
Lazaridou, A., Biliaderis, C. G., Bacandritsos, N., & Sabatini, A. G. (2004). Composition,
thermal and rheological behavior of selected Greek honeys. Journal of Food
Engineering, 64, 9–21.
Muli, E., Munguti, A., & Raina, S. K. (2007). Quality of honey harvested and processed
using traditional methods in rural area of Kenya. Journal of Acta Veterinaria Brno,
76, 315–320.
Nandaa, V., Sarkara, B. C., Sharmaa, H. K., & Bawab, A. S. (2003). Physico-chemical
properties and estimation of mineral content in honey produced from different
plants in Northern India. Journal of Food Composition and Analysis, 16, 613–619.
Nuru A. (1999) Quality State and Grading of Ethiopian Honey. In: Proceeding of 1st
National Conference of Ethiopian Beekeepers Association (pp. 74–82), 7–8 June,
1999, Addis Ababa, Ethiopia.
Oddo, P. L., & Piro, R. (2004). Main European unifloral honeys: Descriptive sheets.
Apidologie, 35, 38–81.
Omafuvbe, B. O., & Akanbi, O. O. (2009). Microbiological and physico-chemical
properties of some commercial Nigerian honey. African Journal of Microbiology
Research, 3, 891–896.
Pridal, A., & Vorlova, L. (2002). Honey and its physical parameters. Czech Journal of
Animal Science, 47, 439–444.
Rehman, S., Khan, Z. F., & Maqbool, T. (2008). Physical and spectroscopic
characterization of Pakistani honey. Cienciae. Invstigation Agraria, 35, 199–204.
Ruoff, K., & Bogdanov, S. (2004). Authenticity of honey and other bee products.
APIACTA, 38, 317–327.
SAS. (2002) SAS Guide to personal computers (Version 9). SAS (statistical analysis
system) institute Inc., NC, USA.
Soria, A. C., Gonzalez, M., Lorenzo, C., Martınez-Castro, I., & Sanz, J. (2004).
Characterization of artisanal honeys from Madrid (Central Spain) on the basis of
their melissopalynological, physicochemical and volatile composition data.
Food Chemistry, 85, 121–130.
Vit, P., Rodríguez-Malaver, A., Roubik, D. W., Moreno, E., Souza, B. A., Sancho, M. T.,
et al. (2009). Expanded parameters to assess the quality of honey from
Venezuelan bees (Apismellifera). Journal of Api Product and Api Medical Science, 1,
72–81.