EUROPEAN COMMISSION

POINT
RESEARCH
i CENTRE
Environment Institute
Ispra (VA) Italy
.5
Evaluation of Methods of Analysis
for Authenticity Proof of Honeys
(Botanical and Geographical Origin)
Elke Anklam
1996 EUR 17274 EN
EUROPEAN COMMISSION
POINT
RESEARCH
! CENTRE
Environment Institute
Ispra (VA) Italy
Evaluation of Methods of Analysis
for Authenticity Proof of Honeys
(Botanical and Geographical Origin)
Elke Anklam
1996 EUR 17274 EN
LEGAL NOTICE
Neither the European Commission nor any person
acting on behalf of the Commission is responsible for the use which
might be made of the following information
Catalogue : CL-NA-17274-EN-C
ECSC-EC-EAEC Brussels Luxembourg, 1996
Printed in Italy
List of Contents
page number
Summary
Introduction . 3
Background 3
Composition 4
Processing 8
Adulteration techniques 8
Objective of this paper 9
Methods of analysis ...... ... ........................................................10
Pollen (Mellisopalynology) 10
Carbohydrates (Sugars) 11
Humidity (Moisture, Water) 14
Nitrogen, Amino acids and Proteins 15
Flavonoids 17
Other Phenolic Compounds 20
Aliphatic Organic Acids 22
Stable Isotopes (Isotopie ratio) 23
Aroma Compounds 25
HMF (5-hydroxymethyl furfural) 27
Fermentation Products 28
Minerals and Trace Elements 28
Enzyme Activity 30
Residues 31
Multi-Component Analysis 34
Special Compounds (Possible Marker Compounds) 37
Conclusion 40
Glossary 42
References 43
Summary
This paper is concerned with analytical methods which allow the
determination of the botanical and geographical origin of honey. The Commission of
the EU has adopted recently a proposal to amend the existing honey Directive in
which it is laid down that the name "honey" has to be supplemented by information
referring to the product's floral and geographical origin. The Commission of the EU is
now encouraging the development of harmonised analytical methods to permit
verification of compliance with the quality specification for the different honey as laid
down in the draft of the new honey Directive. Methods described in the literature are
summarised and evaluated in this paper in order to prove their suitability for detection
of the botanical and geographical origins of honey. The conclusions of this work may
facilitate further analytical work in order to prevent frauds and to protect authentic
honey samples.
Whereas the determination of some single parameters such as 5-
hydroxymethyl furfural (HMF), residues, enzyme activity, moisture, nitrogen and
mono- and disaccharides in honey does not lead to any information about the
botanical and geographical origin there are some suitable methods based on analysis of
specific components or on multicomponent analysis. Mostly, such methods give
indications on the botanical origin investigating flavonoid patterns, distribution of
pollen, aroma compounds and special marker compounds. There are some other
profiles of components which could probably be used for the detection of the
geographical origin (e.g. oligosaccharides, amino acids, trace elements).
In particular, the combination of methods could be a promising approach for
proof of authenticity, especially, when modern statistical data evaluation techniques
will be applied. The most promising method combination could consist of analysis of
trace elements, oligosaccharides, amino acids, flavonoids, aroma compounds and
oxygen stable isotopie ratios in the water of honey. In addition all methods should be
supported by multivariate statistical data analysis.
Introduction
Background
The Council Directive 74/409/EEC concerning honey lays down common rules
for its composition and manufacture. The Commission of the European Union (EU)
has adopted recently a proposal to amend this Directive in which is laid down that the
name "honey" has to be supple-mented by information referring to the product's
floral and geographical origin:
"1. The product names listed in Annex 1 shall apply only to the products referred
to therein and must be used in trade to designate them. These names may be
replaced by the simple product name "honey", except in the case of baker's
honey or industrial honey.
2. Except in the case of industrial or baker's honey, paragraph 1 shall not prevent
these product names from being supplemented by information referring to:
floral or vegetable origin, if the product comes essentially from the
indicated source and possesses its organoleptic, physicochemical and
microscopic characteristics.
regional, territorial or topographical origin, if the product comes
entirely from the indicated source.
specific quality criteria
3. The Member States may provide for the indication of the country of origin for
honeys which do not originate in the European Union."
The Commission of the EU is now encouraging the development of
harmonised analytical methods to permit verification of compliance with the quality
specifications for the different honeys.
Composition of honey
Honey is produced by honey bees from nectar of plants as well as from honey
dew. Some of the components (carbohydrates, water, traces of organic acids,
enzymes, amino acids, pigments, pollen and wax) are due to maturation of honey,
some are added by the bees and some are derived by plants.
Honey from the same floral source can vary due to seasonal climatic variations
or to different geographical origin.
According to the Codex Alimentarius the definition of honey is:
"Natural sweet substance produced by honey bees from nectar or blossoms or from
secretions of living parts of plants or excretions of plant sucking insects on the living
parts of plants, which honey bees collect, transform and combine with specific
substances of their own, store and leave in the honeycomb to ripen and mature".
This definition is valid in all countries which adopt the Codex Alimentarius. There are
additional definitions of honey in the Codex Alimentarius and regulations of various
countries and of the EU [Molan, 1996]:
Blossom honey or nectar honey is distinguished as coming from the nectaries
in flowers
Honeydew honey is distinguished as coming from excretions of insects that
suck the sap of plants
Various physical forms of honey
* Pressed honey
* Centrifiiged honey
* Drained honey
Styles of honey
* Comb honey (whole combs or sections of combs, in its natural state)
* Chunk honey (pieces of comb in honey removed from the comb)
* Crystallised or granulated honey (removed from the comb and having
undergone a natural process of solidification)
* Creamed honey (the glucose crystals in honey were crushed to have a
fine crystalline structure)
* Heat processed honey
Honey immediately after extraction is at its best in terms of flavour and
colour. It is not suitable for large-scale marketing without further treatment. Most
producers sell most of their honey to processors who prepare it for marketing and
package it. As extracted, "raw" honey contains extraneous matter such as pollen,
traces of wax, variable amounts of sugar-tolerant yeasts, and probably crystals of
dextrose hydrate. Unless the moisture content is below 17 %, no fermentation takes
place. Most honey will crystallise in time unless action is taken to prevent it.
Processing of honey thus includes controlled heating to destroy yeasts and dissolve
dextrose crystals, combined with fine straining or pressure filtration [White, 1978].
Different processing methods lead to a liquid or solid product.
Honey consists mostly of the monosaccharides glucose and fructose (85-95 %
of the total carbohydrate content). The actual proportion of glucose to fructose in any
particular honey depends largely on the source of the nectar. The average ratio of
fructose to glucose is 1.2 : 1 [White, 1975 a; 1980 b]. Sucrose is present in honey
about 1 % of its dry weight, however this level can be increased if the beekeeper has
over-fed the bees with sugar during the spring. The amount of reducing disaccharides
is about 6.5 % and that of oligosaccharides about 1.5 % [White, 1975 a].
Crystallisation characteristics of sugars can serve as indicator for the botanical
origin. Honeys having a higher proportion of glucose than fructose (e.g. from rape and
dandelion) crystallise rapidly while honeys with a higher proportion of fructose do
not granulate at all.
The mineral content varies from about 0.04 % in pale honeys to 0.2 % in some
dark honey samples. This content is dependent on the type of soil in which the
original nectar bearing plant was located.
The protein content of honey is normally less than 0.5 %. A small fraction of
the proteins are enzymes, these include: invertase, diastase, glucose oxidase and
catalase.
There are many other minor constituents of honey including very low
concentrations of vitamins and plant acids.
The average composition of honey is listed in Table 1.
Table 1:
Average composition of honey
Component
Water
D-Fructose
D-Glucose
Sucrose
Maltose
Oligosaccharides
Protein
Minerals
Vitamins & amino acids
Average value
in%
17
38
32
1.3
7.3
1.5
0.3
0.2
1.0
. Amongst the compositional criteria prescribed in the existing EU Directive on
honey and in the draft for the new EU Directive are requirements relating to the
concentrations of acidity, apparent reducing sugar (calculated as invert sugar) and
apparent sucrose, 5-hydroxymethyl furfural (HMF), mineral content (ash), moisture
and water-insoluble solids. The levels prescribed are listed in Table 2:
Table 2: Compositional criteria described
Parameter
Acidity
Apparent reducing sugar, calculated as invert sugar
Blossom honey
Honeydew honey and blends
Apparent sucrose content
In general
Honeydew honey and blends
and some special honeys
HMF content
Diastase activity (Schade scale)
In general
Honeys with low natural enzyme content
(e.g. ctirus honeys) and an HMF content
of not more than 15 mg/kg
Mineral (ash) content
In general
Honeydew honey and blends
Moisture content
In general
Heather honey and clover honey
Industrial honey or baker's honey
Water insoluble solids content
In general
Pressed honey
Value
not more than 40 %
not less than 65 %
not less than 60 %
not more than 5 %
not more than 10 %
not more than 40 mg/kg
(ppm)
not less than 8
not less than 3
not more than 0.6 %
not more than 1.2 %
not more than 21 %
not more than 23 %
not more than 25 %
not more than 0.1 %
more than 0.5 %
Processing of honey
A little amount of honey is sold in the form of honeycomb, however, most is
extracted from the comb for marketing. The extraction is carried out in different ways:
Crushing or pressing the comb, then straining to remove wax and other debris
Cutting off the caps from the cells with a knife, then either draining or
centrifuging.
Honey is usually warmed to a temperature of 32-40 C in order to lower the
viscosity. This facilitates the extraction, straining or filtration. This temperature is
similar to that in beehives and does not affect the honey very much during the
relatively short processing period. However, some honey samples are heated to higher
temperatures for liquefaction or pasteurisation reasons.
Adulteration techniques
An extensive overview was recently published [Molan, 1996]. The here listed
adulteration techniques are not related to mislabelling concerning the floral or
geographical origin. They are based on "dilution" of honey:
Water addition:
However, the natural water content of honey varies, regulations have set
standards for the maximum content of water. According to the Codex
Alimentarius and to the EU Directive the maximum content of water
permitted in honey is 21 %, except for clover and heather honeys (23 %).
Extension with sugar and syrups:
Mostly glucose was added to honey in times past [Graham, 1979]. A cheap
form of glucose is com syrup and the latter is a traditional adulterant of honey
[Doner, 1979 a; 1979 b].
In more recent times high fructose com syrup (HFCS) has been used as
favoured extender. HFCS can be obtained from glucose by enzymatically
isomerisation to fructose and is similar to honey.
Bee feeding with sugars and syrup:
This fraud is a result of a careless beekeeping practice. Following
recommendations, bees should not be fed with HFCS or other sugars when
there is any chance that this sugar will be extracted. Another recommendation
is not to feed bees at all with HFCS to avoid this adulteration.
Artificial honey:
Such honey is totally fraudulent. Due to the fact that many phenylacetic
esters possess a honey flavour [Jacobs, 1955] these compounds can be used in
the production of artificial honey. There are some patents describing the
production of such a honey [Sidorciuc, 1972; Morton, 1959].
Objective of this paper
This work is not concerned with frauds based on "dilution" of honeys as
described above. The aim of this paper is to evaluate methods described in the
literature (research methods, official methods) in order to prove their suitability for
the determination of the origin (botanical and geographical) of honeys. The
conclusions of this work may facilitate further analytical work in order to help
implementation of the Directive's amendment and to prevent frauds and to protect
authentic honey samples.
Methods of analysis
Pollen (Mellisopalynology)
Traditionally, the determination of the floral origin of honey has been achieved
by analysis of the pollen present in honey. This method is based on the identification
of pollen by microscopic examination and requires much experience of the analyst.
The different pollen types are described in the literature in order to facilitate the
identification (e.g. Moore, 1978; Louveaux, 1978; D'Albore, 1978; Sawyer, 1988].
However, there are some problems related to this method [Molan, 1996]:
Different plant species produce different proportions of pollen
The amount of pollen can vary from season to season
The nectar yield can be different in male and female flowers
Pollen can be filtered out in the bee's honey sac [Maurizio, 1975 a]
Bees can take pollen without collecting nectar, much of pollen can be collected
from plants that cannot be the sources of honey
Filtering of honey for packing for sale
Straining of honey when strainers are used below 200 (pollen grains are in
5-200 in diameter).
Another limitation of this method for authenticity proof is that pollen can be
added fraudulently.
In the case of citrus honey, pollen analysis is not as useful as in honey of
some other floral origins because the amount of pollen in citrus honey is generally
small and very variable [Serra Bonvehi, 1987].
Naturally produced honeys can never derive from a single botanical source.
The term "unifloral" honey is used to describe honey produced mostly from one plant
species. Generally, the pollen content for a honey to be called "unifloral" should be at
least 45 % of the toal pollen content (Maurizio, 1975). This percentage is not of valid
when a floral source leads to nectar with a higher or lower content of pollen grains
than the average. For example, unifloral chestnut honey requires at least 90 % of the
pollen from Castanea but uniforal citrus honey only needs 10 % of pollen from citrus
[Molan, 1996].
10
Conclusion
Pollen analysis can be used for the identification of the floral origin of honey.
However, the results have to be evaluated carefully due to the fact, that there are
many problems related to this method as described above.
Carbohydrates (Sugars)
Sugars (saccharides) represent the main components of honey. Besides the
two main constituents, the monosaccharides glucose and fructose, there are other
minor components consisting of about 25 oligosaccharides (disaccharides,
trisaccharides, tetrasaccharides).
Limited availability and increased price of honey have provided major
incentives for falsification with other carbohydrate materials. In addition to the
traditional adulterants such as invert syrup and conventional corn syrup (CCS), high
fructose corn syrup (HFCS) is also used for adulteration.
One test for the presence of added invert syrup is the determination of 5-
hydroxymethyl furfural (HMF). This test is somewhat ambiguous, because HMF can
legally be present in honey that has been subjected to heat or abusive storage.
Knowledge of the carbohydrate composition of honey is useful in judging its
authenticity.
Saccharides can be determined by a number of different methods based on the
use of their physical characteristics, such as optically, density and refractive index
measurements or by chemical reactions using reactive groups. Chromatography is
useful for the separation and detection of saccharides.
The contents of sugars (reducing and non reducing) in honey can be determined
by various analytical methods:
* Polarisation method for detection of the monosaccharides, fructose and glucose
and the di saccharide sucrose [White, 1980a]
* Titrimetric method after reduction of Cu(II) to Cu(I) by the reducing sugars
present in honey; a modified method was based on the oxidative titration of
Cu(I) with a standard of N-bromophthalimide or N-bromosaccharin [Kumar,
1988]
* Photometric method for determination of reducing sugars by the neocuproine
method using flow injection analysis [Peris-Tortajada, 1992]
11
* Potentiometrie Stripping analysis (PSA) of the reducing sugars [Nanos, 1991]
* Detection of reducing sugars with a 2,4-dinitrophenolate selective membrane
electrode [Gritzapis, 1989]
* Sucrose in honey was determined by a special hybrid biosensor [Barlikova,
1991] or by special enzyme electrode [Xu, 1989]
* Enzymatic method for analysis of glucose [Schwedt, 1988 & Le Marrec,
1990], fructose [Prado, 1994], glucose, fructose and saccharose [Alamanni,
1994] and of glucose, saccharose and maltose by flow injection analysis
[Tzouwara-Karayanni, 1990]
* Thin layer chromatography of (TLC) of glucose and fructose [Pukl, 1990],
glucose, fructose, maltose, sucrose and ribose [El-Kheir, 1991] or isoglucose
[Allegretti, 1987 & Sangiorgi, 1988] or high performance thin layer
chromatography of glucose and fructose (HPTLC) [Patzsch, 1988]
* Gas chromatography (GC) of various mono-, di- and trisaccharides such as
glucose, fructose, sucrose, maltose, xylose, raffinose and melezitose after
derivatisation [Deifel, 1985; Mateo, 1987; Low, 1988a & Bonvehi, 1989]
* High performance liquid chromatography (HPLC) of fructose, glucose and
sucrose with an refractive index (RI) detector [Cirilli, 1986; Bogdanov, 1988;
Campos, 1989 & Bugner, 1992]
* Ion chromatography (IC) with amperometric pulsed detector (APD) of mono-,
di- and oligosaccharides [Peschet, 1991]
Several authentic Ligurian (Italy) honey samples have been studied with
respect to their sugar composition by GC [Zunin, 1987]. The aim of this study was
to detect addition of syrups to honey. The maltose/isomaltose ratio was shown not be
suitable for the detection of adulteration with syrups. However, the determination of
the sucrose and erlse content was shown to be possibly suitable for this purpose.
The addition of sucrose in concentrations less than 5 % [Lipp, 1989] or the
distinction between authentic honey from honey produced by artificially fed bees
could be detected by HPLC [Calagno, 1987].
The amount of sucrose can decrease by storing due to the presence of the
enzyme invertase [White, 1992].
Analysis of oligosaccharide profiles
The most common syrups used to adulterate honey are invert syrups (IS)
produced from cane or beet sucrose, and high-fructose com syrup (HFCS). These
12
syrups are cheap and their carbohydrate profile can be easily manipulated in order to
resemble that of honey.
At least 12 closely related minor disaccharides and 7 trisaccharides were
identified and quantified by GC after trimethylsilylation [Low, 1988a, 1988b &
1995]. Similar results were described in the literature using also GC methods [Deifel,
1985 & Mateo, 1987].
Anionexchange liquid chromatography was shown to be a suitable tool for
oligosaccharide profile analysis [Swallow, 1994]. Honey, chemically and
enzymatically produced invert syrups and high fructose corn syrup (HFCS) contain a
complex mixture of oligosaccharides which are formed during production processes of
these food. The presence of fingerprint oligosaccharides can be used to detect the
illegal use of HFCS and IS in honey.
The oligosaccharide profiles of 91 authentic English honey samples were
obtained by high performance anionexchange liquid chromatography (HPAE) with
pulsed amperometric detection (PAD)[Goodall, 1995]. Multivariate statistical
techniques applied for investigation of these profiles have shown to be useful tools
for the determination of the botanical source.
Official methods
* Titratric method for reducing sugars [AOAC 920.183], sucrose [AOAC
920.184] and fructose, glucose, sucrose, reducing dicsaccharides as maltose,
higher sugars ("dextrin") after separation by column chromatography [AOAC
954.11; 979.21]
* Liquid chromatographic method for glucose, fructose and sucrose [AOAC
977.20]
* High performance liquid chromatography (HPLC) with RI detection
[Schweizer Lebensmittelbuch, 1995]
* Enzymatic determination of glucose and fructose [Schweizer Lebensmittel
buch, 1995]
* Thin layer chromatography (TLC) for detection of high fructose corn syrup
[AOAC 979.22]
* Carbon ratio mass spectrometric (IRMS) method for detection of com syrup
(including high fructose corn syrup) [AOAC 978.17]
13
Conclusion
Most of the methods described above are suitable for the determination of
various sugars in honey and also for detection of addition of sucrose, high fructose
com syrup and invert syrups.
The analysis of the oligosaccaride profiles (based on GC or HPAE) in combination
with multivariate statistical techniques could be a promising method for the detection
of the botanical origin.
Humidity (Moisture, Water)
The moisture content of honey can be determined by refractometry
[Wedmore, 1955]. The refraction index (RI) is proportional to the density
respectively water content of honey.
Equilibrium relative humidity (ERH) is an important parameter for
determination of the microbial and enzymatic stability of food. Polyols can be used
to estimate the ERH of honey and other food samples [Steele, 1987]. The polyols
glycerol, ethylene glycol and propylene glycol were found to be suitable for ERH by
measuring the refractive index after they had equilibrated in the headspace of food.
The polyol method can be used in countries where access to sophisticated equipment
is limited.
Other methods for water content determination are Karl-Fischer titration and
gravimetry [Zrcher, 1980].
Official methods
Official methods are based on refractometry [DIN, 1992; AOAC 969.38;
Schweizer Lebensmittelbuch, 1995] oron direct drying of the honey samples [AOAC
925.45CorD].
Conclusion
The analysis of the moisture content of honey does not represent an usefull
tool for the authenticity proof of origin, at least being the only method.
14
Nitrogen, Amino Acids and Proteins
The nitrogen content of honey is low and varies. The mean value is 0.04 % (40
mg in 100 g honey) with a high standard deviation [White, 1978a]. About 33-55 % can
be lost by ultrafiltration [Paine, 1934; Bergner, 1975].
The amino acid prolin dominates in honey, representing 50-85 % of the total [White,
1978a].
Honey contains about 0.2 % protein [White, 1978b] which is of bee and plant
origin [Lee, 1985; Stadelmeier, 1986; Croft, 1986]. This includes bee -amylase and
other enzymes [Stadelmeier, 1986].
Nitrogen content
The content of nitrogen has been used to detect frauds in honey samples of
Venezuela. Samples containing less than 10 mg nitrogen per 100 g honey have been
considered to be adulterated with sugars [Vit Olivier, 1987].
Amino acids
It has been suggested that certain ratios between contents of various amino
acids could be used to determine the geographic source of a honey. Davies has
analysed 98 honeys from six different geographical sources. However, also the
botanical origin varied [Davies, 1975, 1976]. Regarding the ratio of aspartic
acid/proline to amides/ phenylalanine it has been shown that there are variations in the
ratios between samples of the same area, however, the variation between sources is
much greater.
Six kinds of honey from different botanical sources (acacia, citrus, chestnut,
rhododendron, rosemary and lime) were analysed by GC and the data were evaluated
statistically in order to gain knowledge about the possible use of amino acid patterns
for classification [Pirini, 1992]. The results have shown that the presence of amino
acids such as arginine, tryptophan, and cystine is characteristic for some honey types.
In some cases, the overall amino acid profile does enable differentiation between
specific types. However, a single amino acid or a group of amino acids could not be
selected being suitable for characterisation of particular kinds of honey.
A number of 44 Spanish honey samples from different botanical origin was
analysed by HPLC [Perez Arquillue, 1987]. The protein amino acids (16) were
determined after acid hydrolysis of isolated protein fractions. Applying discriminant
15
analysis of the results obtained the authors could detect local and botanical differences
with satisfactory results.
17 free amino acids have been determined by GC in 45 samples of honey
obtained from the UK, Australia, Argentina and Canada [Gilbert, 1981]. The results
were analysed statistically (using canonical variate analysis) in order to test the
validity of the use of amino acid data for characterising the geographical origin of
honey. Good discrimination was shown between the samples from Australia,
Argentina and Canada. Those from the UK were classified as a group but they fell
between the samples from Argentina and Canada, and could be wholly discriminated
only on further analysis when the Australian group was omitted. These results have
shown that certain groups of honey samples from certain foreign countries could be
distinguished from one another on the basis of free amino acids.
HPLC methods were used in order to determine the total amounts of proline,
leucine and phenylalanine and their enantiomeric ratios in a variety of different honey
samples [Pawlowska, 1994]. Significant amounts of D-leucine and D-phenylalanine
have been found in honeys of different botanical and geographical origins. The amino
acid showing greatest variability within the samples analysed was leucine. The
authors suggested that the enantiomeric ratios of amino acids could be used to test for
storage effects, age, and the processing technique.
Proteins
SDS-PAGE (Sodium dodecyl sulfate-polyacryl amide gel electrophoresis) and
high-resolution two-dimensional electrophoresis was performed in order to investigate
trace proteins of unconcentrated Australian honeys. The detected protein constituents
were assumed to be predominantly of bee origin instead of floral origin. Only one
sunflower sample showed a protein pattern probably due to pollen origin [Marshall,
1987].
Different protein fractions from Spanish (Galicia) honeys were separated by
electrophoresis in Polyacrylamide gel and 12 different fractions could be observed.
Applying discriminant analysis some classification (with a percentage of 87 % on the
base of two fractions) could be done [Rodriguez Otero, 1990].
Official methods
The determination of the nitrogen content is based on the Kjeldahl method
[AOAC 962.18; 960.52]. Proline, the predominant free amino acid, can be determined
16
spectrophotometrically after reaction with ninhydrine [AOAC 979.20; DIN 10.754,
1993; Schweizer Lebensmittelbuch, 1995].
Conclusion
The determination of the nitrogen content gives only indication of adulteration
with sugars.
The amino acid profiles could give indication of the botanical source of honey
samples. However, using only a single amino acid is not a suitable method for
characterisation of special botanical types. The amino acids arginine, tryptophan, and
cystine have been shown to be characteristic for some floral honey types.
Enantiomeric ratios of amino acids could be used for detection of various different
processing techniques and therefore geographical origin. Significant amounts of D-
leucine and D-phenylalanine have been found in honeys of different botanical and
geographical orgins.
The analysis of the amino acid profiles seems to be more suitable than that of
the protein composition. However, the methods must be employed in conjunction
with other techniques in order to obtain a reliable identification of country of origin.
Flavonoids
Flavonoids are a large family of plant phenolic pigments. Many plant systems
contain an extensive number of flavonoids and each plant tends to have a distinctive
profile.
The flavonoid content reaches about 0.5 % in pollen, 10 % in propolis and
about 6000 ppm in honey. Only flavonoid aglycones seem to be present in propolis
and honey while bee pollen contain flavanols in herosidic forms. The flavonoids in
honey and propolis have been identified as flavanones and flavanones/flavanols
[Riborio Campos, 1990].
The antimicrobially active flavanone pinocembrine was found to be present in
11 of 12 honey samples of different origin [Bogdanov, 1984]. A number of 4 Swiss
honey samples (two of floral, two of honeydew origin) was analysed by HPLC and
the main flavonoid determined was pinocembrine. Its concentration varied between 2
and 3 ppm (200-300 g/100 g honey) [Bogdanov, 1989].
However, flavonoid analysis of honey seems to be a promising technique in
studies of the botanical and geographical origin of honey samples [Amiot, 1989].
17
The flavonoid composition in sunflower honey was analysed by GC [Berhia,
1993]. Among various non identified compounds 6 flavone-flavols and 4 flavanone-
flavols could be determined. The main peak was pinocembrin.
In a similar work the flavonoids of sunflower honey were isolated and
analysed by different methods: TLC, HPLC, GC-MS [Sabatier, 1992]. Five main
components were identified as pinocembrin, pinobanksin, chrysin, galangin and
quercetin. Two minor favonoids were also detected: tectochrysin and kaempferol. The
authors suggested, that such flavonoid structures would provide an index of floral
origin with further study.
Characterisation of flavonoids present in sunflower honey and propolis was
achieved in order to assess the relative effects of different components of honey and
propolis [Siess, 1996]. Honey and propolis contained the same major flavonoids:
pinocembrin, chrysin, galangin and pinobanksin.
Flavonoids present in twenty samples of Portuguese heather honey were
analysed by HPLC [Ferreres, 1994a]. The total amount of flavonoids ranged between
60-500 g/100 g honey. However, in Spanish rosemary honey 500-2000 g/100 g
honey were found. The main flavonoids identified were:
* flavanones: pinocembrin, pinobanksin
* flavones: chrysin, galangin
All samples contained a similar flavonoid partem composed of at least 22 compounds.
For heather honey, the most characteristic substances found were: myricetin,
myricetin-3-methylether, myricetin-3'-methylether and tricetin. These four flavonoids
could not be detected in other floral honeys so far. They could be probably be used as
markers for the botanical origin of heather honey.
Another study by the same authors concerning the flavonoid fraction of nectar
(collected from the honey-stomach of bees gathering nectar from heather flowers in
Portugal) has shown that the following 4 main flavonoids were present [Ferreres,
1996]: quercetin, kaempferol-3-rhamnoside, myricetin-3'-methylether and isorhamne-
tin-3-rhamnoside. Since the natural glycosides are hydrolysed by bee enzymes to
render the corresponding aglycones (metabolites detected in honey), acid hydrolysis
was achieved. The aglycones quercetin, kaempferol, myricetin-3' -methylether and
isorhamnetin were identified, as well as ellagic acid. It has been concluded that ellagic
acid and myricetin-3'-methylether (which have not been identified in any of
monofloral honeys investigated so far) seem to be potential markers for the floral
origin of heather honey.
18
A number of 27 Spanish honey samples from the La Alcarria region has been
analysed by HPLC [Ferreres, 1992]. The total flavonoid content ranged between 500
and 2000 g/100 g honey. The major flavonoids were pinocembrin, pinobanksin and
chrysin. The honey samples were directly provided by the bee-keepers and had not
been industrially processed. However, the botanical origin was not specified. A total
of 18 different flavonoids were detected in the honey samples analysed.
Another study by the same authors has shown that correlation between
botanical origin an flavonoid profiles is possible [Ferreres, 1991]. 10 selected Spanish
samples from the La Alcarria region were analysed (5 rosemary, 2 lavender and 3
multifloral honeys).
Seventeen flavonoid aglycones could be identified in various experimental and
commercial citrus honey samples by HPLC analysis [Ferreres, 1993]. The flavanone
hesperetin could be detected in all citrus honey samples. This flanone was not found
in any of the other honey samples investigated (including rosemary, lavender,
sunflower, almond, sweet chestnut, white clover etc.) These results suggest that
hesperetin could be used as a marker compound for the botanical origin of citrus
honey.
A common flavonoid pattern was observed in the different samples which
leads to the suggestion that pollen is not the main source of honey flavonoids. A close
correlation between the flavonoids of honey and propolis has been found suggesting
that flavonoid analysis could be more useful in geographical origin determinations than
in botanical origin studies. The same authors have developed a simple extraction
technique for honey flavonoid HPLC analysis [Ferreres, 1994b].
By means of micellar electrokinetic capillary chromatography (MECC)
correlations between flavonoid patterns and the botanical origin of various Spanish
honey samples could be established [Ferreres, 1994c]. The analytical conditions have
been applied to honeys from lavender, rosemary, citrus and heather. In citrus honey
an accumulation of hesperetin was found [Ferreres, 1994d]. The main compound in
rosemary was found to be 8-methoxy-kaempferol and that in lavender luteolin. This
MECC study has shown that the flavonoid pattern cannot be used for determination
of the geographical origin. Honey samples from Spain, Mexico and Canada were
analysed using this technique and no significant differences were found.
Capillary electrophoresis (CE) could represent an alternative technique in
honey flavonoid analysis to HPLC [Delgado, 1994].
19
Official methods
No official method is based on the analysis of flavonoids in honey.
Conclusion
Characteristic flavonoid patterns could be determined for honeys of special
botanical origin (e.g. for heather honey, citrus honey, sunflower honey etc.) In
addition, the analysis of flavonoid patterns could be also a possible tool for the
geographical origin. The analysis of flavonoids can be performed by HPLC or CE.
Multivariate statistical data evaluation may improve the suitability of this method
approach.
Other phenolic compounds
During the second metabolism of plants various hydroxybenzoic and
hydroxycinnamic acids are formed [Gross, 1981]. It has been shown that
concentrations of such substances differ in various plants [Herrmann, 1979 & 1989].
Phenolic acids
Aromatic carboxylic acids (phenolic acids) arise from the phenyl-propanoi d
metabolism in plants. Various honeys from different floral sources were analysed
regarding their phenolic acid contents by GC after methylation [Steeg, 1988a]. Rape
honeys were characterised by the occurrence of phenylpropanoic acid and buckwheat
honeys had a higher content of 4-hydroxybenzoic acid and no phenylacetic acid.
Heather honeys could be identified by the presence of a high concentration of benzoic
acid, phenylacetic acid, mandelic acid and b-phenyllactic acid. Differentiation of
honeydew honeys and lower honeys was shown to be possible because of the
difference in the concentration of protocatechuic acid.
Phenolic acids in honey samples have been analysed also by HPLC with
coulometric detection. The concentration in honeys were between 0.01 and 10 ppm
(10-1000 g/100 g honey) [Sontag, 1989]. A buckwheat honey extract has been
analysed and the following compounds could be identified: 3,4-dihydroxybenzoic
acid, 4-hydroxyphenyllactic acid, 2,5-dihydroxybenzoic acid, 4-hydroxyphenlyacetic
acid, 4-hydroxybenzoic acid, 3-hydroxybenzoic acid, 3,4-dihydroxycinnamic acid, 4-
20
hydroxy-3,5-dimethoxybenzoic acid, 2-hydroxybenzoic acid, 4-hydroxycinnamic acid
and 4-hydroxy-3-methoxycinnamic acid.
Applying similar method conditions honeys of various floral types were
compared and characterised by the same authors [Jrg, 1992]. The distribution
pattern of phenolic acids allows to differentiate between honey dew, chestnut and
forest blossom honey.
Phenolic esters
An HPLC method was also applied for analysis of phenolic esters in chestnut,
clover, dandelion, fir, linden, orange, rape, robinia (false acacia) and sunflower honey
[Jrg, 1993]. The characteristic values taken were: methyl-4-hydroxybenzoate,
methyl-vanillate and methyl-syringate.
These compounds could be detected mostly in all honey samples analysed. In robinia
honey only methyl-syringate was found. Rape honey had a very high concentration of
methyl-syringate as well. The content of methyl-4-hydroxybenzoate was higher in
rape and orange honey. Some small differences were observed between the varieties
chesnut, clover, dandelion, linden and sunflower. The results allowed clear
differentiation between rape and robinia honey.
Aromatic carbonyl compounds
Aromatic carbonyl compounds leading to strong flavour are produced in a
secondary enzymatic reaction from the respective phenolic acids such as mentioned
above and are present in honey in minor concentrations [Steeg, 1988b].
Aromatic carbonyl compounds were isolated from solvents extracts of honey
and analysed by GC [Husler, 1990]. Such compounds detected in honey were:
benzaldehyde, phenylacetaldehyde, acetophenone, trans-cinnamic aldehyde, 2-
anisaldehyde, 4-anisaldehyde, vanillin and 3,4-dimethoxy-5-hydroxybenzaldehyde.
There was no indication on the botanical origin given in this investigation. The
same authors have analysed various carbonyl compounds in honey of different floral
sources such as chesnut, acacia, buck wheat, eucalyptus, orange and sunflower
[Husler, 1990]. The phenyl propane metabolites salicylaldehyde, p-toloylaldehyde,
vanillin, 2,5-dimethoxybenzaldehyde and 3,4-dimethoxybenzaldehyde could be
detected as natural occurring minor components (5-180 ppb) and were found to vary
in their concentrations in the various honey samples.
21
Conclusion
A careful evaluation of the patterns concerning phenolic acids, phenolic
estersand aromatic carbonyl compounds could probably give an indication for the
botanical origin of honeys.
Aliphatic organic acids
A number of 32 aliphatic dicarboxylic acids was identified as methylesters in
extracts of four unifloral New Zealand honeys by GC-MS. The constituents 2-
methylbutanediocic acid (O-methylmalicacid) and 4-hydroxy-3-methyl-trans-2-
pentenediocic acid were proposed as floral marker substances for New Zealand
rewarewa (Knightea excelsa) honeys [Wilkins, 1995].
An HPLC method was described in order to characterise organic aliphatic
acids in honey samples after purification by solid phase extraction (SPE) [Cherchi,
1994]. The average recoveries of the acids ranged from 89 % to 104 % and the
detection limits from 0.002 to 3 ppm. Italian honey samples analysed were unifloral
such as strawberry-tree, asphodel, and red gum and their botanical origin was checked
by pollen analysis. Mean concentrations of organic acids in these samples found were:
gluconic acid: 2-12 g/kg = 0.2-1.2 %; pyruvic acid: 9-78 mg/kg (ppm); malic acid:
69-145 mg/kg (ppm); citric acid: 64-160 mg/kg (ppm); succinic acid: 12-48 mg/kg
(ppm); fumarie acid: 0.5 - 2.6 mg/kg (ppm).
Official methods
The organic acids are not specified using official methods. Only the acidity
(free, lactone and total) of honey is determined by using a titrimetric method [AOAC
962.19; DIN 10756, 1995, Schweizer Lebensmittelbuch, 1995].
Conclusion
Only little work is published on the composition of organic acids in various
types of honey. However, investigation of the organic acid profile and pattern
evaluation by statistical methods could be at least helpful for having additional
information on honey samples from various sources.
22
Stable Isotopes (Isotopie ratio)
As already mentioned, honey is often adulterated with the relatively cheap
high fructose com syrup (HFCS). Carbon stable isotopie ratio analysis (SIRA) can be
used to detect honey that has been adulterated with other sugars.
The natural abundances of the stable isotopes of the main bioelements in
biogenic material are submitted to small variations caused by isotopie effects of
physical processing and chemical reactions in the natural cycles of these elements.
Resulting typical relative abundances (values) of food can permit assignments to
their origin and treatment and could represent in some cases the proof of adulteration
(e.g. honey with high fructose com syrup) [Schmidt, 1986 & Croft, 1987].
SIRA has been used for the detection of adulteration with com syrup or cane
sugar for various food (honey, maple syrup, apple juice etc.). Applying this method
small quantities in the
13
C content of the carbon of different plant types produced by
different photosynthetic pathways are measured. Most fruits and grains are Calvin
cycle (C3) pathway plants yielding
13
C values near 25 %o; cane and com are Hatch
Slack (C4) pathway plants with
13
C values near 10 %>.
Using the AOAC official method (honey, maple syrup, apple juice, orange
juice), the sample is combusted with oxygen at 850 C and the combustion gases are
recirculated over hot CuO. The formed CO2 is trapped and purified cryogenically and
L3
C/
12
C ratios are determined by IRMS. Another method is to obtain combustion with
CuO in evacuated sealed glass tubes at 550 C. Slightly differences could be observed
[Krueger, 1993]. The coupling of an elemental analyser with an isotopie ratio mass
spectrometer (IRMS) allows online isotopie measurements [Pichlmayer et al,
1988].The
13
C values of various honey samples analysed were about 23.2 to 24.6
%0 .
As the 5
13
C values alone of honey cannot always be definitely used to prove
adulteration by addition of C4 plant sugars, they have been determined in correlation
to those of the protein from honey. The protein value could be used as an internal
standard. For authentic honey samples, a mean difference of + 0.1 %o (range: + 1.1 %o
to 0.9 %o) has been measured. More negative differences indicate the addition of C4
plant sugars [Romann, 1992 & White, 1992]. The limit for the detection of
adulteration is 7 %. Addition of C3 plant sugars (beet sugar) cannot be proved by this
method. However, for certain types of honeys, authenticity can be confirmed by the
deuterium (D) values. [Romann, 1992].
23
Using the difference in stable carbon isotope ratio (SIRA) between a honey
and its protein fraction an evaluation of honey adulteration with amounts of 7-20 %
and larger of com or cane sugar can be done. A number of 50 authentic honey samples
were used for establishing the purity criteria and 38 other samples with 6
13
C-values in
"questionable" or "adulterated" range were tested. A difference of 1.0 %o or more
between honey and protein fractions was proposed to indicate adulteration [White,
1989].
Carbon-13 nuclear magnetic resonance (
13
C-NMR) was applied for the
qualitative and quantitative analysis of structurally similar disaccharides in honey
[Low, 1988c]. The disaccharides were: glucose-glucose and glucose-fructose.
13
C-
NMR was also applied for the analysis of a complex mixture of minor disaccharides in
honey. Disaccharide ratios (maltose, sucrose, kojibiose, palatinose, turanose,
gentiobiose, neotrehalose, nigerose and isomaltose) in alfa-alfa honey and in sweet
clover honey obtained by
13
C-NMR were compared to those obtained by GC
analysis.
Honey samples from Israel were characterised with respect to the isotopie
ratio parameters 5
13
C, measured by MS and deuterium/hydrogen (D/H) of the methyl
group of the ethanols produced by alcoholic fermentation, measured by deuterium
NMR [Lindner, 1996]. Ethanols obtained from fermentation of citrus honeys have
D/H values similar to ethanols from citrus juice and that exceed the values obtained
from other honeys by 5 ppm. This difference in D/H can be used to confirm the
authenticity of citrus honey. The 5
13
C values of all honeys tested were similar and
typical to C3 plants.
Official methods
The official method is based on carbon ratio mass spectrometry (SIRA)
[AOAC 978.17].
Conclusion
Carbon stable isotopie ratio analysis (SIRA) regarding the 6l3C-values,
especially in combination with correclation to the protein value being used as internal
standard, allows to detect the addition of sugars (com syrup, cane sugar) to honey.
This method is not suitable for the determination of the botanical and geographical
origin.
However, the ratio deuterium/hydrogen (D/H) could represent an useful
method for the determination of citrus honeys. This method could probably be
24
extended to various other floral honey samples and to other stable isotopes in the
samples such as oxygen (
18
0/
16
0). The analysis of the latter ratios could give an
indication of the type of water being present in honey and therefore probably of the
geographical origin.
Aroma compounds
Flavour/fragrance qualities of food products and also of honey are greatly
dependent on the volatile and semivolatile organic compounds present both in the
sample matrix and in the headspace aroma. Volatiles contribute significantly to honey
flavour and to its variation with floral origin and method of handling. Identification of
volatile components is of importance to the understanding of flavour. An elucidation
of the origin of aroma compounds should lead to a better understanding of factors
causing flavour differences between honeys.
The isolation of volatile components from a complex mixture such as honey
and the production of representative extracts is very difficult. In the specific field of
honey, accurate quantification is essential in order to evaluate flavour changes lined to
processing methods or long storage. Such knowledge would further be helpful in
ascertaining a honeys floral origin.
The simultaneous distillation-extraction (SDE) system developed by Likens
and Nickerson [Likens, 1964] and its modified version [Godefroot, 1981] is one of the
most applicable method for the isolation of volatile compounds from a matrix. Since
heat treatment can lead to artefacts, this extraction method has been again modified by
the use of vacuum which leads to isolation of volatile compounds at room temperature
[Maignial, 1992]. Comparing the results of an atmospheric SDE with those of SDE
under vacuum from a non specified honey sample it could be seen that the
atmospheric SDE led to a cooked honey flavour (main components were furfural and
HMF) whereas the latter gave a furfural-free extract with a fresh honey note. The
corresponding GC exhibited a small peak of linalool oxide [Maignial, 1992].
A commercial Canadian honey was used for the optimisation of the Likens-
Nickerson method [Bouseta, 1995]. Dichloromethane extraction under an inert
atmosphere followed by simultaneous steam distillation-dichloromethane extraction
appeared to be a useful method for honey flavour characterisation. A complex mixture
of hydrocarbons, alcohols, phenols, ethers, aldehydes, ketones, esters, furans and
nitrogen compounds could be isolated and identified. The following 19 constituents
25
could be identified in the GC/MS of the Canadian honey sample: methylfuran,
caproaldehyde, octane, furfural, furfuryl alcohol, 1hexanol, mxylene, acetylfuran,
HMF, benzaldehyde, apinene, phenol, pinene, benzyl alcohol, phenylacetal
dehyde, phenethyl alcohol, camphor, coumarin, transcaryophyllene.
Volatile organic compounds were purged from nine commercially available
honeys from various floral sources (wildflower, blueberrry, orange, clover, tpelo,
alfalfa, apple spread, mixed and a mild cream geographical origin not indicated)
followed by trapping on an adsorbent resin [Overton, 1994]. The adsorbent traps
were subsequently analysed by thermal desorptionGCMS. This technique
permitted the analysis of a wider range of both, volatile and semivolatile organic
compounds and was shown to be more sensitive by a factor of at least 100 as
compared to the headspace technique. The honey samples analysed were found to
contain numerous mono and sequiterpenoid compounds and flavours such as
benzaldehyde, furfural, isovaleraldehyde, and phenyllactaldehyde. The presence of
the branched aldehydes methylbutyraldehyde and 3methylbutryraldehyde in each
of the honeys reflected the microbial quality and thermal treatment of honey. The
author stated that this technique used together with pollen analysis may be used in
floral source verification.
The volatile components of a unifloral Italian chestnut honey were isolated by
steam distillation extraction and investigated by GC/MS [Bonaga, 1986a]. Linear
hydrocarbons, saturated and unsaturated, at even and odd numbers of carbon atoms,
from CIO to C37 were found in chestnut honey. nHeptacosane, nnoncosane, n
tricosane, npentacosane, and nhentriacontane were the largest gas chromatographic
peaks in the alkane fraction of volatiles (about 40 %), whereas ntritriacontene and
nhentriacontene predominated in the unsaturated portion (about 60 %). The
positional and geometrical isomerism of the double bond in alkenes was investigated
by the study of their epoxides.
The same authors have shown that the extract of volatiles from Italian
chestnut honey was a complex mixture of at least 50 compounds of which some of
these, and particularly the main component of the mixture (3aminoacetophenone)
may be specific for the floral source [Bonaga, 1986b].
The flavour compound l(2,6,6trimethyll,3cyclohexadienyl)2butenlone
(damascenone) was quantified by a stable isotope dilution assay (SEDA) method in
two honey samples (geographical origin not indicated) [Grosch, 1990]. The content
was about 3 ppb (\ig/kg) in acacia honey and about 8 ppb in linden honey.
26
Official methods
There are no officials methods described for flavour analysis of honey.
Conclusion
The analysis results of the aroma of honey can vary with the isolation and
detection mode of the flavour constituents. The honey flavour is dependent on the
method of handling (processing, storage etc.) and botanical source. A careful analysis
of the volatiles in honey being responsible for the aroma could be a usefool tool for
characterisation of the botanical source. Some typical compounds could be identified
for honeys from special floral sources (e.g. chestnut honey). The methods on the
aroma profile evaluation should be combined with other techniques.
HMF (5-hydroxymethyl furfural)
Low levels of 5-(hydroxymethyl)-2-furfuraldehyde = 5-hydroxymethyl
furfural (HMF) are usually detected in honey due to the degradation of sugars
catalysed by the normal honey acidity at room temperature. This degradation is
accelerated during heat processing or storage at high temperatures. The occurrence of
high levels of HMF in honey may also be caused by adulteration with commercial
invert syrup [White, 1980b].
The HMF content in honeys allows an evaluation of the honey quality. The
limit set by the EU and Codex Alimentarius Commission Standards is 40 ppm
[Codex Alimentarius Commission Standards, 1981].
Analysis of HMF in honey was carried out by HPLC on a reversed phase
column with UV detection at 280 nm [Vinas, 1992; Marini, 1985 & Jeuring, 1980].
Other methods applied were ion-exclusion chromatography (IEC) with ultra violet
(UV) detection (detection limit 50 ppb) [Kim, 1992] and spectrophotometry [White,
1979a; Espinosa-Mansilla, 1993 & Wood, 1993].
Official methods
Official methods are based on the spectrophotometry determination of HMF
in honey. [AOAC 980.23, Schweizer Lebensmittelbuch, 1995].
27
Conclusion
The analysis of the HMF content gives only an indication about the
processing (heat treatment) and storage of honey but is not suitable for determination
of the botanical nor geographical origin.
Fermentation products
Glycerol occurs as a minor constituent in honey and is probably produced by
micro-organisms present in the nectar and honeydew which are collected by the bees.
Glycerol may therefore be considered as a fermentation product. In the fermentation
of a 20 % glucose solution, aeration and low phosphate content favour the production
of polyols such as glycerol, whereas anaerobic fermentation produces mainly ethanol.
The glycerol content of 33 honey samples from Galicia (Span) has been
determined using an enzymatic method ranging between 50 and 370 mg/kg (ppm)
[Huidobro, 1993].
Primary alcohols of 33 unpasteurised Galicain (Spain) honeys have been
determined as apparent ethanol contents by means of an enzymatic method
[Huidobro, 1994]. The apparent contents of the samples analysed were in the range
from 14 to 50 ppm.
Official methods
There are no official methods described in the literature.
Conclusion
The content of fermentation products such as glycerol and ethanol in honey
samples is dependent on micro-organisms present in honeydew and nectar and gives
information about the processing of honey (pasteurisation). This approach is not
suitable for authenticity proof of honey.
Minerals and trace elements
The contents of minerals and trace elements in honey samples could give an
indication of the environmental pollution and herewith also an indication of the
geographical origin of honey.
28
The minerals sodium, potassium, calcium, magnesium, copper, iron,
manganese, phosphorus (phosphate), chlorine (chloride), silicon (silica), sulphur
(sulphate) and ash contents of 91 samples of raw honey from Galicia (Spain; floral
source not indicated) were determined [Rodriguez-Otero, 1994]. Potassium was
shown to be the most abundant of the elements determined, with an average content
of 1500 mg/kg (ppm). In general, the honey samples investigated have shown higher
mineral contents in comparison with honeys reported in the literature. The same
authors have analysed 24 commercial Spanish honey samples concerning their silicon,
phosphorus, sulphur, chlorine and ash contents [Rodriguez-Otero, 1995]. Mean
contents were 3 ppm for silicon, 80 for phosphorus, 45 for sulphur and 260 for
chlorine. The values for phosphorus and chlorine are high with respect to honeys
from other regions.
An ion chromatographic (IC) method was applied for the analysis of inorganic
anions (chloride, hydrogenphosphate and sulphate) in Spanish honey (not specified)
[Perez-Cerrada, 1989].
Trace elements in Italian heather honey samples have been analysed by
preseparation neutron activation analysis (PNAA) [Pietra, 1993].
Instrumental neutron activation analysis was also applied to the determination
of various elements in Turkish honey samples from various floral origins (mixed
flower, sunflower, thyme and citrus flower) [Sevimli, 1992].
The trace elements lead, cadmium and manganese in honey samples (floral
source not indicated) from various seasonal origin (spring, summer, autumn) were
analysed with minimal sample preparation (dilution with water) by graphite-furnace
atomic absorption spectrometry (ETAAS) [Stein, 1986]. The lead contents were
found to be slightly higher in the summer honey samples.
The elements selenium, iron and calcium were also determined by ETAAS in
honey samples (origin not specified) [Dabeka, 1991, Szymosyk, 1986 & Siong, 1989a
and 1989b].
Trace element concentrations in honey collected from a wide area in Hungary
were shown tobe useful as an environmental indicator [Fodor, 1993]. The efficiency
and reproducibility of sample preparation (dilution, ashing and digestion) for
simultaneous multi-element analysis have been compared for 13 elements in honey. In
almost all cases the trace element concentrations in honey samples from industrial
areas were higher than in those from relatively clean areas. This was shown to hold
true especially for the elements cadmium, copper, lead and zinc.
29
Official methods
The ash content of honey is obtained by the classical furnace method [AOAC
920.181, Schweizer Lebensmittelbuch, 1995]. There is no official method available
concerning the mineral or trace element contents.
Conclusion
The analysis of the contents of minerals and trace elements in honey could be
suitable for detection of the geographical origin due to the fact that these values are
affected very much by the environmental pollution. The investigation of trace element
profiles in combination with modem statistical data evaluation techniques could be a
promising approach.
Enzyme activity
The enzyme activity could be a measure for exposure of honey to heat in
processing and storage. However, this value is less exact than that of the HMF
content because enzyme activities vary a lot for various honey samples. This is due to
the fact that different amounts of saliva containing enzymes can be added by the bees
to honey under different conditions.
The activity of the enzyme diastase in honey is related to its heat treatment.
The diastase is also called amylase or -amylase. The diastase activity can be
determined by reaction with starch and iodine as indicator by spectrophotometry
[Hadorn, 1972; Bogdanov, 1984].
Official methods
The official methods are based on the same principal as described above
[AOAC 958.09; DIN 10750, 1990; Schweizer Lebensmittelbuch, 1995].
Conclusion
The diastase activity gives only an indication about the processing (heat
treatment) of honey samples but is not suitable for the detection of the origin. The
detection of the activity is always limited by the variability in composition.
30
Residues
Deriving from medicinal treatment of honey bees
The insecticide/acaricide amitraz is used also against Varroa Jakobsoni of
honey bees. Both, amitraz and its transformation product in honey, 2,4-xylidine, can
be detected by HPTLC [Pavoni, 1993]. This method is only valid for qualitative
determination.
An HPLC and an ELISA method were developed in order to detect fumagillin,
having antibiotic properties for winter honey bee medication. The detection level of
the HPLC method was 100 ppb, that of the ELISA method 20 ppb [Assil, 1991]. At
these detection limits there was no evidence of fumagillin or its breakdown products
found in any Canadian producer honey examined. The investigations included also
samples from a beekeeper who regularly used fumagillin.
1,4-Dichlorobenzene is used as an acaricidial preparation against Varroa
Jakobsoni. The residues were determined by head space analysis in honey [Binder,
1988]. The detection limit is 0.1 ppm. Nearly no degradation of 1,4-dichlorobenzene
in contaminated honey being filled in glass bottles for commercialisation could be
detected.
Sulphonamides are used for prevention and treatment of diseases in honey
bees. Sulfamonomethoxine could be detected by HPLC in 2 of 87 commercial honey
samples in concentrations of 0.2 respectively 0.8 ppm [Horie, 1992}. (35 from Japan,
47 from other origin not specified
Cymiazole is also used as acaricide against Varroa Jacobsoni. The detection
limit of an HPLC method has been shown to be 0.01 ppm [Carras, 1993]. No
indication was given about the contamination of commercial honey samples.
Four acaricides (amitraz, fluvalinate, malathion and terramicine) against Varroa
Jacobsoni were analysed in 35 Italian honey samples [Saita, 1993]. In 4 samples
fluvalinate (0.5-92 ppb) and in 1 sample malathione (1 ppm) were found. The
formation of possible metabolites in honey seems to be a problem.
Antibiotics
The antibiotic Oxytetracycline, controlling both American and European
foulbrood (destroying honeybee larvae) was analysed by HPLC [Sporns, 1986]. This
compound was determined to be more stable in honey than in buffered aqueous
solutions at similar pH values and temperatures.
31
The HPLC method developed for analysis of residues of tetracycline
derivatives in honey was shown to have a sensitivity of 0.1 ppm for the qualitative
and 1.0 ppm for the quantitative analysis [Jrgens, 1981].
The HPLC method for analysis of residues of chloramphenicol and
sulfathiazole allows complete recovery and has a sensitivity of 1 ppm [Jrgens,
1982].
An analytical system for the simultaneous determination of residual
Oxytetracycline, tetracycline, Chlortetracycline, doxycycline, methacycline,
demethylchlortetracycline and minocycline in honey has been established by a
combination of simple TLC and HPLC methods [Oka, 1987a]. In this system,
screening by TLC can detect tetracyclines at a level of 0.1 ppm in honey and the
quantitative HPLC method can determine such compounds with good recovery. No
indications about the detection limit of HPLC have been given. The same authors have
improved the sample preparation method for HPLC using a tandem cartridge clean-up
system [Oka, 1987b]. The detection limits in honey are 0.02 ppm for Oxytetracycline
and tetracycline, and 0.05 ppm for Chlortetracycline and doxycycline, respectively.
Derivative spectrophotometrical methods have been developed for
determination of Oxytetracycline and doxycycline in honey without any pre-treatment
of the samples [Salinas, 1989]. A similar method was also applied for the
determination of sulphonamides in honey and other biological samples [Salinas, 1990].
Menthol, used as treatment against Acarpi s woodi in bee hives and found in
honey samples was not reduced by storage in open containers at room temperature
for up to 55 days [Li et al, 1993]. The levels in honey were found to reach about 40
ppm.
Repellents
Phenol is widely used as a repellent to drive bees from the honeycomb for
honey collection. Phenol can be determined by HPLC with amperometrical detection
[Taskeba, 1990] or by UV detection [Sporns, 1981].
A number of 112 Japanese honey samples was analysed [Taskeba, 1990]. In 32
samples phenol could be detected. The range and average values were 0.05 - 5.88 ppm
and 0.71 ppm, respectively. The detection limit of phenol in honey was found to be
0.002 ppm.
An analytical method for the analysis of various bee repellents residues to 1
ppm in honey samples was developed based on GC [Kwan, 1988]. The compounds
analysed were: butyric acid from butyric anhydride, propionic acid from propionic
32
anhydride, phenol, and benzoic acid from benzaldehyde. From field trials it seemed
that the use of phenol and propionic anhydride left considerably larger amounts of
residue in honey than the use of butyric anhydride or benzaldehyde (most Canadian
honey samples analysed containing only trace levels).
Pesticides
Spanish citrus honey samples from Valencia were analysed by GC-ECD in
order to determine the content of the organochlorine pesticides such as lindane, aldrin,
dieldrin, endrin and ,'-DDT [Serra Bonvehi, 1985]. Positive samples showed the
following concentration ranges: lindane 0.07-0.12 ppb, aldrin 0.05-0.07 ppb, dieldrin
0.05-0.10 ppb and endrin 0.05 ppb. The DDT isomer could not be detected.
A similar method with a modified sample preparation was developed for the
detection and quantitative determination of organochlorine pesticides in honey
(lindane, heptachlor, aldrin, heptachlor epoxide, dieldrin, endrin and methoxychlor).
Others
The toxic compounds for humans, tutin and hyenanchin, could be detected in
honey samples of certain areas of New Zealand [Love, 1986]. An HPLC method has
been developed. Most of the samples from New Zealand were negative, but one
contained 2 ppm tutin and 3 ppm hyenanchin.
Literature on analytical work on residues of the acaricides amitraz,
brompropylate, coumaphos, cymiazole, fluvalinate, malathione and phenothiazine
between 1981 and 1994 was reviewed [Femandez-Muino, 1995].
Official methods
There are many official methods described in the literature concerning the
analysis of the various pesticides used. However, they are not only dedicated to the
matrix honey and therefore not mentioned in this paper.
Conclusion
Due to the fact that many pesticides are used widespread and also in various
countries they can be find in many pollen and therefore the analysis of them in honey
samples is not very indicative neither for geographical nor for botanical origin. The
same holds true for resues deriving from medicinal treatment.
33
Multi-component analysis
Compliance with legal aspects
A collaborative trial concerning the analysis of various parameters as
prescribed in the existing EU Honey Directive was carried out in order to evaluate the
suitability of the analytical methods [Lord, 1988]. The results indicated that the
proposed methods of analysis for the determination of mineral content, moisture,
acidity, apparent reducing sugar content and water-insoluble solids content were
satisfactory while those for HMF and apparent content required further investigation
at that time investigated. The values determined were not used in terms of floral or
geographical authenticity proof.
Methods for determination of carbohydrates, HMF and proline were
established and evaluated [White, 1979b].
The adulteration of honey samples with high-fructose corn syrup (HFCS) was
detected by applying different methods [Abdel-Aal, 1993]. Pure honey was
adulterated with HFCS with levels of 10 % to 50 %. The sugar composition as a
fingerprint was determined by HPLC. The following compositional properties were
determined in addition for pure and adulterated honey samples: moisture, total soluble
solids, nitrogen, apparent viscosity, HMF, ash, sodium, calcium, potassium, proline,
refractive index and diastase activity. Statistical analysis revealed that the following
compositional properties were highly significantly negatively correlated with sugar
composition, dry matter, apparent viscosity, sodium, potassium, proline and
nitrogen. In contrast, ash, calcium, HMF and moisture were highly significantly
positively correlated with sugar composition of pure and adulterated honey.
The water activity (a,) of various Italian honey samples (prunus, sunflower,
acacia and orange) was determined together with other parameters such as sugar and
water contents and the physical characteristics such as fluid or crystallised status
[Piana, 1991]. The various honey types could be characterised by their different
aw/water content values.
Italian honeys from the market were tested concerning their physical, chemical
and organoleptic characteristics [Bolchi Serini, 1981]. Parameters such as refraction
index, colour, viscosity, impurity, reductive sugars, acids, ash, enzymes and HMF
were determined, but only with respect to the legal aspects and not for use of control
of origin.
34
The carbohydrates glucose, fructose and sucrose and 2 organic acids (malic and
citric acid) in two Italian flower honeys were analysed by enzymatic methods [Tourn,
1980].
A number of 11 parameters for honey quality was determined in 29 honey
samples of Spain [Lopez, 1996]. Multivariate chemometric techniques such as
principal component analysis, cluster analysis and linear discriminant analysis were
used to classify honey samples on the basis of the chemical data. Using the values of
total acidity and diastase activity a mostly correct classification for natural authentic
honeys to processed honeys could be achieved. However, no indications on the floral
or geographical origin have been made.
The contents of sugars and minerals in Spanish honey were determined by
polarimetrie, reductometric, HPLC and enzymatic methods [Frias, 1992].
The physicochemical characteristics of 25 samples of commercial Spanish
eucalyptus honey were analysed and 35 parameters were measured, including
contents of sugars and minerals, total nitrogen, proline, water content, pH, acidity,
HMF, diastase activity, colour, ash, insoluble solids and electrical conductivity
[Gomez, 1993]. The samples contained a mean of 18 pollen types. Samples with
more than 70 % of eucalyptus pollen were considered to be unifloral. In 92 % of the
samples analysed, eucalyptus was the most abundant pollen, at over 45 % of total
pollen found. The quality of 27 rosemary honey samples from Spain was evaluated
[Perez-Arquillue, 1994]. Most samples showed a proper maturity considering the low
moisture content. The low electrical conductivity and ash content were typical of pale
honeys. Other parameters analysed were the contents of sucrose, glucose, fructose,
trisaccharides and HMF.
Unifloral Spanish honey samples (willow, sainfoin, chickweed, crucifer,
fruiter, thyme, blueweed, lavender and vetch) were analysed regarding their moisture
content, optical rotation, electrical conductivity, ash and HMF contents, diastase
activity, pH, acidity and carbohydrate composition [Perez-Arquillue, 1995]. The
samples were considered to be unifloral when more than 45 % of pollen of the
respective flowers were found. No classification on and no evaluation of these
parameters was done in view to the floral origin of the samples analysed. These
samples were just found to meet the major national and international honey
specifications.
Cluster analysis was applied to physical chemical parameters obtained from
Spanish honeys from the Basque country [Sancho, 1991]. The parameters have been
joined in 5 clusters: free acidity, total acidity, fructose content, glucose content and
35
diastase activity. Proline content and total reducing sugars content could also be
formed into clusters. The samples themselves have also been joined into 2 clusters:
one cluster had only 1 sample which had the higher diastase activity.
The adulteration of Spanish honey samples with com syrup was proven by
analysis of various parameters such as HMF content, isomaltose/malose ratio,
moisture content, pH, ashes, sulphur dioxide, viscosity and density [Serra Bonvehi,
1986].
Based on a comparative study of analysis of the humidity , the ashes and the
sugars in a sample of Portuguese honey, a recommendation is made for the
introduction of analysis methods in the Portuguese legislation concerning honey [Pena
Ferreira, 1989].
Origin control
Geographical
Honey samples from two production areas in the north-western part of Spain
(Galicia) were analysed according to 11 parameters of legal quality control [Pena
Crecente, 1993]. The samples were from mixed floral origins from the 1990 harvest.
Classification of these honeys according to their geographical origin was achieved by
pattern recognition techniques to the chemical data. Humidity and free acidity were
found to be the most important parameters for the classification.
Authentic honey samples from the province of Alberta, Canada, were
analysed regarding their moisture, enzymes (a- and -glucosidase, diastase), HMF,
proline, free and lactone acidity, ash and carbohydrate composition in order to
establish original values [Sporns, 1992]. The samples were found to meet all major
national and international honey specifications. By comparison of the results to those
obtained from other world floral honeys, these samples were found to be lower in
contents of moisture, a-glucosidase, proline, acidity and ash. The proline levels were
often lower than 200 ppm which has been used as a minimum level for honey
authenticity by some Canadian honey importers.
A number of 37 honeys of Swiss and other geographical origin were analysed
concerning the contents of trace insecticides, water, protein, sugars, HMF and
activities of water (a
w
), diastase, saccharase and heat-stable inhibines and the light
absorption, colour intensity and electric conductance [Bogdanov, 1987]. No detectable
amounts of the insecticides fumidile and sulfathiazol could be detected. There was no
correlation between inhibines and the botanical (honey-dew or floral) and the
geographical (Swiss or foreign) honey origin. Swiss honey samples had significantly
36
less water, lower water activity, higher activities of saccharase and diastase and a
lower content of HMF than the foreign honeys.
Botanical
A number of 18 chemical and physical parameters of nectar and honeydew
honeys were determined and the results were analysed statistically by the method of
principal component analysis (PCA) [Krauze, 1991]. The honeys could be divided
into the following groups: acacia, rape, linden and heather, and honeydew. The most
important first principal component was strongly associated with the value of
electrical conductivity, the contents of ash, free acids and proline, as well as with the
pH and the diastase number. The principal component loadings and linear correlation
suggested that these parameters contributed much more to the classification of honeys
than apparent reducing sugars, apparent sucrose, mono-, di- and trisaccharides,
glucose and fructose.
The palynological (pollen) and physicochemical properties of 15 citrus honey
samples commercially produced in Spain were reported [Serra Bonvehi, 1995]. Sugar
profiles and the flavour component methyl anthranilate change were shown to change
with the transition period before honey commercialisation. Therefore the
recommendation was made to classify fresh Spanish citrus honey with the group of
abundant nectariferous flower honey in order to admit its higher sucrose content
which was shown to exceed largely the legal limits established for honey by the EU.
Conclusion
As demonstrated by the examples given above an approach using various
methods and data evaluation with modem statistical techniques could be promising in
order to detect the botanical and geographical origin of honeys. The parameters taken
into account have to be chosen very carefully and should be limited to a minimal
number in order to keep the laboratory costs low but to obtain optimal results.
Special compounds (possible marker compounds)
Abscisic acid in heather honey
The HPLC analysis of Portuguese heather honey fractions have shown that
two organic acids (non-flavonoid compounds) were the main constituents [Ferreres,
1996]. These compounds were isolated and identified as cis,trans-abscisic acid and
37
trans,trans-abscisic acid. Their contents ranged between 0.3 and 17 ppm in honey.
These compounds were not detected in any of the different monofloral honey samples
analysed so far, and therefore, could be useful markers of heather honey. More
heather honey samples from other geographical and botanical origins should be
analysed in order to prove that abscisic acid could be a useful marker of heather
honey.
(S)-(+)-dehydrovomifoliol in heather honey
This odourless compound was found in levels of 190-260 pg/g (ppm) in four
French heather honeys investigated and of 60 ppm in a Spanish heather honey
sample. Other honeys from other floral sources contained this compound only in a
concentration of 0.03-6 ppm [Husler, 1989].
Hexenyl butyrate in eucalyptus honey
Various Australian honeys have been investigated by GC-MS and only in
eucalyptus honey the compound hexenyl butyrate could be detected [Graddon, 1979].
Lindenether (3,9-epoxy-I,4[8]-p-menthadiene in linden honey
Cis-rose oxde in linden honey
Trans-8-p-menthen-l,2-diol in linden honey
These terpenes have been detected exclusively in linden honey but not in any
other honey sample investigated (robinia, orange blossom, rape, wild tobacco) [Blank,
1989].
Hesperetin in citrus honey
The flavanone hesperetin was suggested as a possible marker for the floral
origin of citrus honey, since this compound had not been detected in honey of any
other origin so far [Ferreres, 1993]. It is a constitutive phenolic compound of citrus
nectar, where it was present as a glycoside (hesperidine).
Methyl anthranilate in citrus honey
Methyl anthranilate is a characteristic volatile component of citrus nectar and
honey and has been used as marker in citrus honey performing the analysis by GC or
spectrophotometrically [White, 1965; Wooton, 1978; Graddon, 1979; Biechi, 1983 &
Serra Bonvehi, 1988].
An HPLC method was described which allows the simultaneous determination
of HMF and methyl anthranilate in honey samples [Vinas, 1992].
38
However, as methyl anthranilate is a volatile compound, it suffers significant
changes in concentration with different environmental factors and under different
honey storage conditions (White, 1964; Wooton, 1978 & Serra Bonvehi, 1988].
The contents of hesperetin and methyl anthranilate were determined by GC
and HPLC respectively in 18 honey samples produced in Mediterranean Spain
[Ferreres, 1994d]. There was no correlation between the contents of both compounds
found. The concentration of methyl anthranilate ranged between 1.4 and 3.6 ppm
while hesperetin ranged between 0.3 and 0.9 ppm. These results support the
hypothesis that hesperetin could be used as an additional marker in the determination
of citrus honey origin.
3-Aminoacetophenone in chestnut honey
This volatile compound was found to be the main constituent of the complex
mixture of volatiles in (Italian) chestnut honey and could be specific for the floral
source [Bonaga, 1986].
Conclusion
The analysis of special marker compounds is facilitating the detection of the
origin. However, not for all botanical varieties there are such special compounds
available or if so many often only in very small quantities. In addition, the
quantification of such components could raise problems due to the fact, that honey is
a natural product and the concentrations can vary.
Some of these markers compounds are easily available and could therefore be
fraudulently added to honey.
39
Conclusion
The suitability of the methods for detection of the botanical and/or
geographical origin of honey as described above is summarised in Table 3.
Table 3: Evaluation of methods and their suitability for detection of the
botanical and geographical origin of honey
Analysis of
Mono- & Di-
saccharides
Oligosaccharides
Moisture
Nitrogen
Amino acids
Proteins
Flavonoids
Phenolic acids
Phenolic esters
Arom. Carb. Comp
Technique
GC
TLC
HPLC
Conductivity
Refiactomey
Titrimetry
GC
Reftactometry
Gravimetry
GC
HPLC
SDS-PAGE
GC
HPLC
MECC
GC
HPLC
HPLC
GC
Botan.
Ongin
(+)
(+)
(+)
(+)
+
+
+
(+)
(+)
+
+
Geogr.
Origin
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Other purposes
Syrup addition
Syrup addition
Syrup addition
IC & HFCS addition
Sugar addition
(+ = suitable, (+) = possibly suitable)
Table 3 will be continued on the next page
40
Continuation of Table 3:
Analysis of
Aliph. Org. Acids
Acidity
SiRA
D/H
Minerals,Trace
elements
Residues
Enzyme activity
HMF
Special marker
compounds
Aroma comp.
Multicomponents
Technique
GC
Titrimetric
NMR
IRMS
NMR
AAS
ETAAS
ICP-MS
HPLC
ELISA
Spectrophotom.
Spectrophotom.
Spectrophotom.
HPLC
Various methods
SDE
GC-MS
SIDA
Various techniques
Botan.
Origin
(+)
+ (citrus)
+
+
+
+
+
Geogr.
Origin
(+)
(+)
+
(+)
(+)
(+)
+
Other purposes
Sugar addition
Heat treatment
Invert sugar addition
Heat treatment
Many other purposes
(+ = suitable, (+) = possibly suitable)
Whereas the determination of some single parameters (HMF, residues,
enzyme activity, SIRA, moisture, nitrogen, mono- and disaccharides) in honey does
not lead to any information about the botanical and/or geographical origin there are
some methods of analysis being suitable for this purpose as listed in Table 3. Most of
the latter give indications on the botanical origin.
The combination of methods (multicomponent analysis) seems to be a
promising approach for resolving the problem of authenticity proof regarding the
origin of honeys. This is especially the case when the most suitable methods (Table 3)
will be combined and modem statistical evaluation techniques such as principal
component analysis (PCA) and neural nets will be applied.
41
Glossary
AAS
APD
a
w
CCS
CE
ELISA
ERH
GC
GC-MS
HFCS
HMF
HPAE
HPLC
HPTLC
IC
IEC
IRMS
IS
MECC
MS
NMR
PCA
PNAA
Ppb
ppm
PSA
RI
SDE
SDS-PAGE
SIDA
SIRA
SPE
TLC
UV
Atomic absorption spectrometry
Amperometric pulsed detector
Water activity
Conventional com syrup
Capillary electrophoresis
Enzyme linked immunosorbent assay
Equilibrium relative humidity
Gas chromatography
Gas chromatography-mass spectrometry
High fructose com syrup
5-Hydroxymethyl furfural
High performance anion exchange
High performance liquid chromatography
High performance thin layer chromatography
Ion chromatography
Ion exclusion chromatography
Isotopie ratio mass spectrometry
Invert syrup
Micellar electrokinetic capillary chromatography
Mass spectrometry
Nuclear magnetic resonance
Principal component analysis
Preseparation neutron activation analysis
Parts per billion tg/kg)
Parts per million (mg per kg)
Potentiometrie stripping analysis
Refraction index
Simultaneous distillation extraction
Sodium dodecyl sulfate-polyacryl amide gel electrophoresis
Stable isotope dilution assay
Carbon stable isotopie ratio
Solid phase extraction
Thin layer chromatography
Ultra violet
42
References
Alamanni, M.C., rv. Sci. Aliment. 23, 8186 (1994)
Allegretti, M., Ambrosoli, G., Cantoni, C, Ind. Aliment. 26, 566573 (1987)
Amiot, M.J., Aubert, S., Gnnet, M. Tacchini, M., Apidologie 20, 115 (1989)
AOAC, Official methods for analysis, Ed. Helrich, K., 15th Ed. (1990)
Assil, H.I., Spoms, P., J. Agrie. Food Chem. 39, 2206 (1991)
Barlikova, ., Svorc, J., Miertus, S., Anal. Chim. Acta 247, 8387 (1991)
Berahia, T., Cenati, C, Sabatier, S., Amiot, M.J., Sci. Ahm. 13, 1524 (1993)
Bergner, K.G., Diemair, S, . Lebensm. Unters. Forsch. 157, 713 (1975)
Biechi, D., Belliardo, F., Frattini, C, J. Apic Res. 22, 130136 (1983)
Binder, H, Krainer, W., Brestschlo, J., Z. Lebensm. Unters. Forsch. 186, 223334 (1988)
Blank, J., Fischer, K.H, Grosch, W., Lebensm. Unters. Forsch. 189, 426429 (1989)
Bogdanov, S., Lebensm. Wiss.u.Technol. 17, 7476 (1984)
Bogdanov, S., Mitt. Geb. Lebensm. Hyg. 75, 214220 (1984)
Bogdanov, S., Apidologie 18, 267278 (1987)
Bogdanvov, S., Baumann, E.,Mitt. Geb. Lebensm. Hyg. 79, 198206 (1988)
Bogdanov, S., J. Apicul. Res. 28, 5557 (1989)
Bolchi Serini, G, L 'apicol. Mod. 72, 1522 (1981)
Bonaga, G, Giumanini, A.G., Gliozzi, G, J. Agrie. Food Chem. 34, 319 (1986)
Bonaga, G, Giumanini, A.G., J. Apicult. Res. 25, 113120 (1986b)
Bouseta, ., Collin, S., J. Agrie. Food Chem. 43, 18901897 (1995)
Bugner, E., Feinberg, M., J. AOAC Internat. 75, 443464 (1992)
Calcagno, C, Zunin, P., Evangelisti, F., Riv. Soc. hal. Sci. Aliment. 16, 453458 (1987)
Campos, M., Da Graca, R., Rev. Port. Farm. 39, 101103 (1989)
Carras, P., Melis, M., Spanedda, L., J. AOAC Internat. 76, 9294 (1993)
Cherchi, ., Spanedda, L. Tuberoso, C. Cabras, P., J. Chromatogr. 669, 59 64 (1994)
43
Cirilli, G., Adana Cirilli, CS. , Pulga, C, Zaghini, L., Ind. Aliment. 25, 3537 (1986)
Codex Alimentarius Commision Standards, Volume III (1981), Codex Standard for Sugars,
Standard 121901, Codex Standard for Honey, FAO Rome
Croft, L.R., Mistry, R.P. and Washington, R.J., in Electrophoresis ' 86' (Dunn, M.J., Ed.),
pp. 338, VCH Publishers, Deerfield Beach, FL (1986)
Croft, L.R., Trends in Anal. Chem. 6, 206209 (1987)
Dabeka, R.W., McKenzie, A.D., Can. J. Appi. Spectr. 36, 123126 (1991)
D'Albore, G.R, Oddo, L.P. "Flora apistica italiana", Federzione apicoltori italiani
(1978)
Davies, A.M.C., J. Apic Res. 14, 2939 (1975)
Davies, A.M.C., J. Food Technol. 11, 515523 (1976)
Deifel, ., Dtsch. Lebensm. Rdsch. 81, 209212 (1985)
Delgado, C, TomasBarberan, F., Talou, T., Gaset, ., Chromatographia 38, 7178
(1994)
Deutsches Institut fr Normung, DIN Norm 10752 (1992)
Deutsches Institut fur Normung, DIN Norm 10750 (1990)
Doner, L.W., Kushnir, I., White, J.W., Anal. Chem. 51, 224A232A (1979a)
Doner, L.W., White, J.W., Philipps, J.G., J.AOAC 62, 186189 (1979b)
ElKheir, A.A., Metwally, M.M.F., Ayad, M.M., Egypt. J. Food Sei. 19, 4151 (1991)
EspinosaMansilla, ., Muoz de la Pena, ., Salinas, F., JAOAC Internat. 76, 12551261
(1993)
FernandezMuinol, M.A., Sancho, M.T., Muniategui, S., Huidobro, J.F., SimalLozano, J.,
J. Food Protect. 58, 449454 (1995)
Ferreres, F., TomasBarberan, F.A., Gil, M.I., TomasLorente, F., J. Sci. Food Agrie. 56,
4956(1991)
Ferreres, F., Ortiz, ., Silva, C, GarciaViguera, C, TomasBarberan, F.A., Tomas
Lorente, F., Lebensm. Unters. Forsch. 194, 139143 (1992)
Ferreres, F., GariciaViguera, C, TomasLorente F., , TomasBarberan,F.A., J. Sei. Food
Agrie. 61, 121123 (1993)
Ferreres, F., Andrade, P., TomasBarberan, F.A., . Lebensm. Unters. Forsch. 199, 3237
(1994a)
Ferreres, F., TomasBarberan, FA., Soler, C. GarciaViguera, C. Ortiz, ., Tomas
Lorente, F., Apidologie 25, 2130 (1994b)
44
Ferreres, F., Amparo Blazquez, ., Gil, M.I., TomasBarberan, F.A., J. Chromatogr.
A669, 268274 (1994c)
Ferreres, F., Giner, J.M., TomasBarberan, F.A., J. Sci. Food Agrie. 65, 371 (1994d)
Ferreres, F., Andrada, P., I. Gil, M., TomasBarberan, F.A., Lebensm. Unters. Forsch.
202, 4044 (1996)
Ferreres, F., Andrade, P., TomasBarberan, F., J. Agrie. Food Chem. 44, 20532056
(1996)
Fodor, P., Molnar, E., Mikrochim. Acta 112, 113 (1993)
Frias, I., Hardisson, ., Alimentaria 225, 4143 (1992)
Godefroot, M., Sandra, P., Verzele, M., J. Chromatogr. 203, 325 (1981)
Gomes, M.E.M., Hernandez, E.G. Gomez, J.Y.M., Marin, J.L.M., J. Apicult. Res. 32, 121
126 (1993)
Goodall, I., Dennis, J.J. Parker, I, Sharman, M., J. Chromatogr. A706, 353359 (1995)
Graddon, A.D., Morrison S.D., Smith J.F., J. Agrie. Food Chem. 27, 832837 (1979)
Graham, J., Amer. Bee J. 119, 228229 (1979)
Gritzapis, P., TimotheouPotamia, M., Anal. Chim. Acta 218, 3746 (1989)
Grosch, W., Sen, ., Guth, H, ZeilerHilgart, G, Flav. Sci. Technol. 6Th Weurman
Symp., p. 191194 (1990)
Gross, G.G., "The Biochemistry of Plants", Vol. 7, Secondary Plant Products, Academic
Press, New York (1981)
Hadom, H, Zrcher, ., Dtsch. Lebensm. Rdsch. 68, 209216 (1972)
Husler, M., Montag, ., Dtsch. Lebensm. Rdsch. 86, 171174 (1990)
Husler, M., Montag, ., Lebensm. Unters. Forsch. 189, 113114 (1989)
Herrmann, K., Lebensmittechem. Gerichtl. Chem. 33, 119 (1979)
Herrmann, K., Crit. Rev. Food Sei. Nutr. 28, 315 (1989)
Horie, M., Saito, ., Nose, ., Nakazawa, ., J. AOAC Internat. 75, 786789 (1992)
Huidobro, J.F., Estrella, R. M., Branquinho de Andrade, P.C., Sancho, M.T., Muniategui,
S., SimalLozano, J., J. Agrie. Food Chem. 41, 557559 (1993)
Huidobro, J.F., Estrella, Rea, M., Branquinho de Andrade P.C.,, Sanchez, M.P., Sancho,
M.T., Muniategui, S., SimalLozano, J., J. Agrie. Food Chem. 42, 19751978 (1994)
Jacobs, M.B., Amer. Perf. Essent. Oil Rev. 66, 4647 (1955)
45
Jeuring, HJ., Kuppers, F.J.E.M., J.AOAC 63, 12151218 (1980)
Jrg, E., Sontag, G., Dtsch. Lebensm. Rdsch. 88, 179183 (1992)
Jrg, E., Sontag, G, J. Chromatogr. 635, 137 (1993)
Jrgens, U., Lebensm. Unters. Forsch. 173, 356358 (1981)
Jrgens, U., Lebensm. Unters. Forsch. 174, 208210 (1982)
Kim, H.J., Richardson, M., J. Chromatogr. 593, 153156 (1992)
Krauze, ., Zalewski, R, . Lebensm. Unters. Forsch. 192, 1923 (1991)
Krueger, D.A., J.AOAC Internat. 76, 418420 (1993)
Kumar, KG, Das, CM., Indrasenan, P., Talanta 35, 651652 (1988)
Kwan, S., Sporns P., J. Apicult. Res. 27, 162 (1988)
Lee, C.Y., Smith, N.L. Kime, R.W., Morse, RA., J. Apic Res. 24, 190194 (1985)
Le Maree, J.H, Lesgards, G, Ann. Fals. Exp. Chim. 83, 393400 (1990)
Li, M., Nelson, D l , Sporns, P., J.AOAC Internat. 76, 1289 (1993)
Likens, ST., Nickerson, GB., ASBC Proceed, p. 5 (1964)
Lindner, P., Bermann, E., Gamarnik, B., J. Agrie. Food Chem. 44, 139140 (1996)
Lipp, J., Ziegler, H., DE 3723190A1 (1989)
Lopez, B., Latorre, M.J. Fernandez, M.I., Garcia, M.A., Garcia, S., Herrero, C, Food
Chem. 55, 281287 (1996)
Lord, D.W., Scotter, M.J., Whittaker, A.D., Wood, R, J. Assoc. Pubi. Analysts 26, 5176
(1988)
Louveaux, J., Maurizio, ., Vorwohl, G., Bee World 59, 139157 (1978)
Love, J.L., Swallow, W.H,, Maister, S.G., N.Z.J. Technol. 2, 179182 (1986)
Low, N.H., Spoms, P., J. Food Sci. 53, 558561 (1988a)
Low, N.H., Nelson, D.L., Spoms, P., J. Apicult. Res. 27, 245251 (1988b)
Low, N.H., Brisbane, T., Bigam, G, Spoms, P., J. Agrie. Food Chem. 36, 953957
(1988c)
Low, N.H., South, W., J.AOAC Internat. 78, 12101218 (1995)
Maignial, L., Pibrot, P., Bonetti, G, Chaintrau, ., Marion, J.P., J. Chromatogr. 606,
8794 (1992)
46
Marini, S., Righi, G, Ind. Aliment. 39, 693 - 694 (1985)
Marshall, T., Williams, K.M., Anal. Biochem. 167, 301-303 (1987)
Mateo, R., Bosch, G, Pastor, ., Jimenez, M.,J. Chromatogr. 410, 319-328 (1987)
Maurizio, ., in "Honey: A comprehensive survey", Ed. E. Crane, Heinemann, London,
p. 77-105 (1975)
Molan, P.C., in "Food authentication", Ed. P.R. Ashurst, M.J. ennis, Chapman & Hall,
London (1996)
Moore, P.D. Webb, J.A., in "An illustrated guide to pollen analysis", Hodder and
Stoughton, London (1978)
Morton, I.D., Sharpies, E., US Patent No 2916382 (1959)
Nanos, CG, Karayannis, M.I., Fres. J Anal. Chem. 340, 253-257 (1991)
Oka, H, Ikai, Y., Kawamura, N., Uno, K, Yamada, M., J. Chromatogr. 400, 253
(1987a)
Oka, H., Ikai, Y., Kawamura, N., Uno, K., Yamada, M., Harada, K., Uchiyama, M.,
Asukabe, H., Mori, Y., Suzuki, M., J. Chromatogr. 389, 417-426 (1987)
Overton, S.V., Manura, J.J., Americ. Laborat. 26, 45-53 (1994)
Paine, H.S., Gertler, S.I., Lothrop, R.E., Ind. Eng. Chem. 26, 73-81 (1934)
Patzsch, K., Netz, S., Funk, W., J. Planar Chromatogr. 1, 177-179 (1988)
Pavoni, G., Boll. Chim. Igten. 44, 283-287 (1993)
Pawlowska, M., Armstrong, D.W., Chirality, 270 (1994)
Pena Crecente, R, Herrero Latorre, C, J. Agrie. Food Chem. 41, 560-564 (1993)
Perez Arquillue, C, Herrera Marteache, ., Alimentaria 24, 67-71 (1987)
Perez Arquillue, C, Conchello, P., Arino, ., Juan, T., Herrerea ., Food Chem. 51, 207
(1994)
Perez Arquillue, C, Conchello, P., Arono, ., Juan, T., Herrera, ., Food Chem. 54,
167-172 (1995)
Perez-Cerrada, M., Herrero-Villen, M.A., Maquieira, ., Food Chem. 34, 285-294
(1989)
Peris-Tortajada, M., Puchades, R., Maquieira, ., Food Chem. 43, 65-69 (1992)
Peschet, J.L., Giacalone, ., Ind. Aliment. Agrie. 108, 583-586 (1991)
47
Piana, ML, Poda, G. Cesaroni, D., Chetti, L, Bucci, M.A., Gotti, P., Riv. Soc. hal. Sci.
Aliment. 20, 273279 (1991)
Pichlmayer, F., Blochberger, K., Fres. . Anal. Chem. 331, 196201 (1988)
Pietra, R, Fortaner, S., Sabbioni, E., J. Trace & Micropr. Techn. 11, 235250 (1993)
Pirini, ., Conte, L.S., Francioso, O. Lercker, G.,J. High Resol. Chromatogr. 15, 165
170 (1992)
Prado, F.E., Sampietro, A.R., Biotechnol. Appi. Biochem. 19, 361368 (1994)
Pukl, M., Prosek, M., J. Planar Chromatogr. 3, 173176 (1990)
Riborio Campos, M.G., Sabatier, S., Amiot, M.J.,Aubert, S., Planta Med. 56, 580581
(1990)
RodriguezOtero, J.L., Paseiro, P. Simal, J., An. Bromatol. 42, 8398 (1990)
RodriguezOtero, J.L., Paseiro, P., Simal, J., Cepeda, Food Chem. 49, 169171 (1994)
RodriguezOtero, J.L., Paseiro, P., Simal, J., Terradillos, L., Cepeda, ., . Lebensm.
Unters. Forsch. 200, 233234 (1995)
Romann, ., Lllmann, C, Schmidt, H.L., Lebensm. Unters. Forsch. 195, 307311
(1992)
Sabatier, S., Amiot, M.J., Tacchini, M., Aubert, S., J Food Sci. 57, 773777 (1992)
Saita, M., Dugo, G, Di Bella G, Leuzzi, U., Visco, A.M., Riv. Seien. Aliment. 22, 8389
(1993)
Salinas, F., Berzas Nevado, J.J., Espinosa, ., Analyst 114, 11411145 (1989)
Salinas, F., Msilla, Espijosa Mansilla, ., Berzas Nevado, J.J., Anal. Chim. Acta 233, 289
294 (1990)
Sancho, M.T., Muniategui, S., Cancela, R, Huidobro, J.F., Simal, J., Anal. Bromatol. 43,
267273 (1992)
Sangiorgi, E., Ind. Aliment. 27, 442444 (1988)
Sawyer, R.W., 'Honey Identification", Cardiff Academic Press, Cardiff (1988)
Schmidt, H.L., Fres. . Anal. Chem. 324, 760766 (1986)
Schwedt, G., Hauck, M., Dtsch. Lebensm. Rdsch. 84, 8285 (1988)
Schweizer Lebensmittelbuch, 5. Auflage (1995)
Serra Bonvehi, J., Alimentaria 22, 5558 (1985)
Serra Bonvehi, J., Pajuelo, A.G., Alimentac. Y Technol., 143147 (1986)
48
Serra Bonvehi, J., GomezPajuelo, ., GonellGalindo, F., Alimentaria 185, 6184 (1987)
Serra Bonvehi, J., Alimentaria 197, 3740 (1988)
Serra Bonvehi, J., Bosch Callis, J., An. Quim. 85, 3846 (1989)
Serra Bonvehi, J. Ventora Coll, F., J. Agrie. Food Chem. 43, 20532057 (1995)
Sevlimli, H, Bayulgen, N, Varinlioglu, ., Radioanal. Nuc. Chem. Lett. 165, 319325
(1992)
Sidorciuc, D., Pospescu, N, Ruman. Patent No 53959 (1972)
Siess, M.H, Le Bon, A.M., CanivencLavier, M.C, Amiot, M.J. Sabatier, S., Yaubert,
S.Y., Suschetet, M., J. Agrie. Food Chem. 44, 22972301 (1996)
Siong, T.E., Choo, K.W. Shahid, S.M., Pertanika 12, 303311 (1989)
Siong, T.E., Choo, K.W. Shahid, S.M., Pertanika 12, 313322 (1989)
Sontag, G, Friedrich, O., Kainz, G. Jrg, E., Agrie. Food Chem. Consum, Proc. EUR 5th
Food Chem. Conf. 1989, Voi 2, pp. 703707
Spoms, P., J. AOAC 64, 337339 (1981)
Spoms, P., Kwan, S., Roth, L.A., J. Food Protect. 49, 383388 (1986)
Spoms, P., Pilhak, L., Friedrich, J., Food Res. Internat. 25, 93100 (1992)
Stadelmeier, M., Bergner, K.G, Lebensm. Unters. Forsch. 182, 196199 (1986)
Steeg, E., Montag, ., . Lebensm. Unters. Forsch. 187, 115 (1988a)
Steeg, E., Montag, ., Dtsch. Lebensm.-Rdsch. 84, 147 (1988b)
Steele, R.J., Internat. J. Food Sei. Technol. 22, 377 (1987)
Stein, K, Umland, F., Fres. . Anal. Chem. 323, 176177 (1986)
Swallow, K.W., N.HLow, J. AOAC Internat. 77, 695702
Szymozyk, S., Kajfosz, J., Dutkienwicz, E., Borkowski, J., Spurenelem. Symp. 5th, p.
598604 (1986)
Takeba, K., Matsumoto, M., Shida, Y., Nakazawa, H, J. AOAC 73, 602604 (1990)
Toum, ML., Lombard, ., Belliardo, F., Buffa, M., J. Apicult. Res. 19, 144146 (1980)
TzouwaraKarayanni, S.M., Crouch, S.R., Food Chem. 35, 109116 (1990)
Vinas, P., Campillo, ., Hernandez Cordoba, M. Candela, M.E., Food Chem. 44, 6772
(1992)
Vit Olivier, P., Acta dent. Venzolana38, 511512 (1987)
49
Wedmore, E., Beeworld 36, 197-206 (1955)
White, J.W., Kushnir, I., Subers, M.H., Food Technol. 18, 555-558 (1964)
White, J.W., J. Food Sci. 31, 162-164 (1965)
White, J.W., in "Honey: A comprehensive survey", Ed. E. Crane, Heinemann, London,
p. 157-206 (1975)
White, J.W., Adv. Food Res. 4, 287-374 (1978a)
White, J.W., J. Apic. Res. 17, 234-238 (1978b)
White, J.W., J.AOAC 62, 509-514 (1979a)
White, J.W., J.AOAC 62, 515-526 (1979b)
White, J.W., J. AOAC 63, 11-18 (1980a)
White, J.W., Siciliano, J., J. AOAC 63, 7-11 (1980b)
White, J.W., Winters, K., J.AOAC 72, 907-911 (1989)
White, J.W., J.AOAC Internat. 75, 543-548 (1992)
Wilkins, L.A., Lu, Y., Tan, S.-T., J. Agrie. Food Chem. 43, 3021 (1995)
Wood, R, J. Assoc. Pubi. Analysts 28, 195-199 (1993)
Xu, Y., Guilbault, G.G., Anal. Chem. 61, 782-784 (1989)
Zrcher, K., Hadom, H., Mitt. Geb. Lebensmittelunters. 71, 396-401 (1980)
Zunin, P., Calcagno, C, Evangelisti, F., Riv. Soc. hal. Sci. Aliment. 16, 317-322 (1987)
50
CLNA17274ENC