elissopalynology is the branch of palynology which studies the botanical and geographical origin of honey by subjecting honey sediment, and therefore pollen and the other fungi imperfecti contained therein, to microscopic analysis.
It should be mentioned that, according to law no. 753/1982, which assimilated the relevant European regulations, honey is "an alimentary product that honey bees produce from flower nectar, from secretions produced by living parts of plants and from excretions found on the plants. This mixture is gathered, transformed, combined with specific substances peculiar to bees, stored and brought to ripeness in the honeycombs of the beehive". It should also be mentioned that there is nectar honey and honeydew honey and that there are unifloral (where one nectar source prevails) and multifloral (diverse nectar sources are present) types of honey.

The earliest research on the pollen analysis of honey was undertaken by Pfister in 1895; since then various other researchers have devoted themselves to this subject (Fehlman, 1911; Armbruster, 1929, 1934-35; Griebel, 1931). The most authoritative of these researchers is certainly Zander, whose works (1935, 1937, 1941, 1949, 1951) are still the main reference point for whoever is interested in this subject.

These studies have made it possible to ascertain, on a rigorous, scientific basis, the apicultural importance of the different botanical species, whereas previously this evaluation was the fruit of general field observations. Even though melissopalynological analysis is not error proof it does provide, together with organoleptic and physical - chemical analysis, a valid means for formulating an objective opinion about the botanical origin of any type of honey. Flower nectar always contains greater or lesser quantities of pollen and this pollen can then be traced in the honey sediment. Identification of these pollens, estimation of the percentage in which they are present and the eventual identification of elements probably indicative of honeydew, make it possible to trace the botanical species gathered with far greater precision than can be obtained with direct observations. With direct observation it is only possible to ascertain whether a species is more or less intensely visited by bees, but not the extent to which it contributes to honey production.

Through melissopalynology it is also possible to trace the geographical origin of a particular type of honey, since its pollen spectrum, i.e. pollens in the sediment as a whole, reflects the floral situation of the place where that particular honey was produced. Different geographical areas present particular floral associations and the greater the climatic difference the more conspicuous the variation in the floral association. The pollen spectrum of a tropical honey is quite different from that of a Mediterranean honey; even varieties of honey produced in areas close to each other or with a similar climate present differences: the rare pollens present in honey vary as does the percentage value of each pollen present. In Italy this phenomenon is rather marked, as phytosociological realities can vary greatly even over short distances; the reason for this lies in the abundance of microclimates created by the orography of the land, depending on our country orography. The identification of geographic origin is generally based upon the presence of a combination of pollens typical of that particular area: only in some cases is it possible to find particular pollens, which are characteristic of a certain territory and are not found elsewhere; these marker pollens, if present, are sufficient to indicate the origin of the variety of honey in which they are found. It would therefore seem feasible to utilize certificates of origin to distinguish among the various different types of honey. This fact is of great practical importance, especially in those countries where laws protecting national beekeeping and honey products are in force, because it prevents dealers from purchasing foreign honey at a low price and then passing it off as nationally-produced honey and selling it at an increased prize, to the obvious detriment of local beekeepers.

A microscopic examination of honey brings to light possible impurities, such as insect fragments, dust, etc. The presence of these substances in honey is forbidden by the laws regulating the sale of this product.

It should be noted that the pollen content of honey can be influenced by numerous factors, some relating to the morphological characteristics of flowers and pollens, and others relating to the operations successively carried out on nectar and on honey.

The nectar can be contaminated by the pollen at various times. There are three pollution categories: primary, secondary and tertiary.

Primary pollution occurs in the flower as a result of the mechanical action of insects, wind, etc. These agents shake the anthers and the pollen becomes detached and falls into the nectar of the same flower. The quantity of pollen which falls into the nectar varies. Both the shape and the position of the flower can facilitate or limit the extent of this pollution; for example, the larger the pollen granules are, the less likely they are to be found in the nectar. Pollen content can also be limited by the following factors: the presence of extrafloral nectars, the lack of synchronism between anther deiscence and the moment of maximum nectar secretion, partial or total sterility of the stamina, unisexuality of the species (female flowers do not contribute to pollution). All the elements responsible for primary pollution are closely related to plant characteristics and are relatively constant in any one single species so it is possible to evaluate them fairly accurately; if these elements taken together lead to an abundance of pollen in the nectar, the pollen is classified as hyperrepresented; on the contrary, if these factors obstruct pollution, the pollen is classified as hyporepresented; in intermediate cases the pollen is classified as normally represented.

Secondary pollution takes place from the moment when the nectar arrives in the hive to the moment when the cell, overflowing with honey, is capped. However, it should be mentioned that some alterations to the pollen content take place during the transport of the nectar to the hive. During the gatherer re-entry flight the nectar is filtered by the proventricular valve which retains a part of the pollen. The longer the nectar remains in the crop the better it is filtered. This filtering operation is more effective for the larger pollens, so their number, already inferior to that of the smaller pollens, is reduced even more. In the hive, during nectar transit from one bee to another and later, while the cells are being filled up, the nectar and honey are enriched with pollen adhering to the bees' hair; this pollen can come from either the nectaripherous species gathered or from the stores of pollen, used to feed the young bees and larvae. The stronger the pollen gathering and hive activities, the greater this kind of pollution. It mainly affects the anemophilous pollens, which are less sticky and more easily dispersed than the entomophilous ones. This secondary pollution, therefore, although less easily verifiable than the primary type, can be partly revealed using microscopic analysis.

Lastly, there is tertiary pollution which takes place during honey-extraction operations and is caused by pollen reserves stored in the hive, especially in the honeycomb, and by pollen dispersed on the honeycomb surface. This pollution is negligible if honey is obtained by centrifugation and if simple sanitary measures, such as washing the honeycombs with warm water before decapping and not drawing brood honeycombs for honey-extraction, are observed. There is also quaternary pollution, caused by the pollens present in the atmosphere, but this phenomenon is much more limited than the preceding ones.

From what has been said, it is clear that the results of melissopalynological analysis, however reliable, cannot guarantee absolute precision. In conclusion, melissopalynology, like other sciences investigating biological phenomena, where it is often difficult to evaluate the variables, can make no claim to be a mathematical science.

The following pages describe qualitative and quantitative methods of subjecting honey to microscopic analysis. These methods are ratified by The International Commission for Plant-Bee Relationship (ICPBR). These methods can be applied to all varieties of honey, except for those filtered with Diatomee sand or soil (a technique frequently used in the U.S.A.) (Ricciardelli D'Albore and Persano Oddo, 1978).

Qualitative melissopalynological analysis

This kind of analysis consists in recognizing the different fungi imperfecti contained in the sediment and in evaluating the respective percentages of each element. In most cases this is sufficient to determine the botanical and geographical origin of honey. It is impossible to discover the botanical origin of honey obtained by pressing, because the sediment is enriched with the contents of the pollen cells. Caution is needed when identifying the botanical origin of Calluna honey because the extraction technique used for this honey results in an abundant sediment. 10 g of honey is dissolved in 20 ml of water at 40 C and then centrifuged for 5' at 2,500 rpm and settled; the sediment is recovered in 10 ml of distilled water and is centrifuged again and then settled; the sediment is then collected with a Pasteur pipette and set onto an object slide where it dries at 40 C; it is then included in glycerinated jelly, covered with a microscope slide and luted. If the type of honey to be analysed contains many colloids it is better to use acidulous water with 5g/l of sulphuric acid, rather than pure distilled water. For the rarely used acidulytic method, reference is made to the specific textbook (Louveaux et alii, 1978).

The next step is the microscopic analysis of the compound and identification of its component elements with the help of reference compounds. The following elements are counted and classified separately:

  • the pollens of nectaripherous plants;
  • the pollens of nectarless anemophilous and entomophilous plants (Papaver, Galega, Thalictrum, etc.);
  • the elements indicative of honeydew: hyphae or fungi spores present in the atmosphere which develop on honeydews, algae, waxy secretions coming from some honeydew-producing insects; these elements should be noted individually, and the pluricellular bodies counted as single units (hyphae, spores complexes, etc.);
  • aborted or misshapen pollens when identification is not possible.

The presence of that fine homogeneous crystalline mass characteristic of honeydew honey, but sometimes found in the honey of some nectaripherous plants (in particular, Ericaceae) should be noted if found.

When identifying pollens it is often not possible to ascertain the exact species and sometimes even the genus is uncertain; in these cases more general categories, based on groups, shapes, and types are used.

The number of granules examined depends on the degree of precision required.
For an indicative sample evaluation the computation of about 100 PK (PK = pollen grain) should be sufficient. It is necessary to count 200-300 PK to determine the frequency classes.
The following nomenclature is used when determining the frequency classes:

very frequent pollen>45%
frequent pollen16-45%
isolated pollen4-15%
rare pollen<3%

For a precise percentage calculation 1000-1200 PK have to be counted (Vergeron, 1964) and the following terms are adopted:

predominant pollen >45%
accompanying pollen16-45%
important isolated pollen4-15%
isolated pollen <3%

Melissopalynologists have recently agreed on the methods to be used for defining the exact percentage of every pollen found in honey sediment. The frequency of honeydew indicators is estimated on the basis of the ratio of these indicators to the pollen grains of nectaripherous plants. Honey is considered honeydew honey if this ratio is greater than 3. In some cases, however, this ratio can be as low as 1:1 (fresh honeydew gathered only a few days beforehand).

A nectar honey is considered to be of a certain species if the pollen of this species exceeds 45% of the total. If there is no predominant pollen then this kind of honey is classified as multifloral.

However, qualitative analysis results cannot always be directly interpreted in this way. The relation between the percentage of a certain pollen and the presence of the corresponding nectar is valid for normal pollens, but it has to be modified for underrepresented and overrepresented pollens. This is because, in the case of underrepresented pollens, the quantity of nectar actually participating in honey formation is superior to what one would have expected from the pollen count, and in the case of overrepresented pollens it is less. Myosotis honey, for example, is so strongly overrepresented that its sediment has to contain 100% species pollen before it can be considered unifloral. Castanea honey has to contain more than 90%, etc.
In honey coming from species with underrepresented pollen, uniflorality is guaranteed by a percentage inferior to the 45% necessary for normal honeys; Lavandula honey is considered unifloral if it contains 5-10% of the species pollen and the same applies to Tilia honey; for Robinia honey 10-20% species pollen is necessary, etc. A particular case is represented by Citrus, whose cultivars are steril and fertil; in America, where cultivars are mainly steril, Citrus honey is considered unifloral if 10% of the pollen is Citrus, but Citrus honey from Calabria can contain more than 60% Citrus pollen.

It should also be remembered that under- and overrepresented honey varieties have a total pollen content which is, respectively, inferior and superior to those of normal honeys. As a consequence, it is necessary to carry out sample quantitative analysis as well, in order to confirm the diagnosis. If there is a correlation between the low percentage of a underrepresented pollen and low absolute contents, then this kind of honey can be considered unifloral; if, however, the absolute contents are high, this type of honey is likely to be multifloral and further analysis investigation is necessary.

In conclusion, in order to define a honey variety as unifloral, it is necessary to know the characteristic percentage of that particular species pollen and the respective absolute pollen value. These values, which are not fixed but fluctuate within certain limits, are obtained by the quantitative and qualitative analysis of numerous samples whose origins are known. If such data are not available it is better to use the term prevalence rather than uniflorality (Ricciardelli D'Albore and Persano Oddo, 1978; Louveaux et alii, 1978).

Quantitative melissopalynological analysis

This kind of analysis involves the evaluation of two different parameters: the total sediment volume and the quantity of fungi imperfecti per honey weight unit.

The determination of the total quantity of sediment per weight unit makes it possible to ascertain how a type of honey was produced and whether or not it contains any foreign matter; it can also be useful in identifying adulterated honey. The method used is as follows: 10 g of honey are dissolved in 20 ml of water at 40 C and centrifuged for 10'; the surnatant liquid is carefully sucked up leaving only 1-2 ml which is then shaken and decanted in a graduated centrifuge tube of the right size, so as to recover all the sediment; it is then centrifuged again for 10' and the sediment volume read off the graduated tube.

A honey variety extracted by centrifugation contains about 1.5-3.5 µl of sediment per 10 g. If there is little or no sediment, this either means that the honey has been filtered with sand or Diatomee soil or that it has been adulterated. The cause of the adulteration is either the direct addition of sugar to the honey or feeding the bees with nectar syrups. If sediment exceeds 10 µl then the honey has either been obtained by pressing or it contains an excessive quantity of solid particles; some of these solid particles may come from the nectar itself (for example, calcium oxalate crystals which are typical of lime-tree, chestnut and mint honey) but a significant amount is certainly due to substances quite extraneous to honey formation. The presence of wax fragments, dust, animal and vegetal hair, portions of insects, etc. is a sign of inadequate hygiene during honey processing; the presence of aleuron granules either means the bees have been fed with too many pollen substitutes or they have been fed at an inappropriate time; lastly, a large number of yeasts is indicative of fermentation.

The determination of the absolute number of fungi imperfecti per unit of honey weight consents a more precise interpretation of qualitative analysis results in the case of honey varieties with under- and overrepresented pollen. This is especially the case when both types of pollens are present in the sediment; there are three different methods for determining the absolute number of fungi imperfecti (Ricciardelli D'Albore and Persano Oddo, 1978; Louveaux et alii, 1978).

Maurizio method

Two samples of homogenized honey, each of 50 g, are weighed, then distilled water is added to a total of 100 ml and the mixture is dissolved in a bain-marie at 40 C. The preparation is then centrifuged for 5' and carefully decanted so as not to lose any of the sediment. The sediment is then shaken to dissolve it in the remaining liquid, and decanted in a 10 ml graduated centrifuge tube. The first tube is then rinsed with a little distilled water and this water is then poured into the 10 ml tube to avoid losing any of the sediment. The preparation is centrifuged a second time for 5' and decanted (or the surnatant liquid is carefully sucked up) and a few drops of distilled water are added so as to bring the volume to a fixed dilution (normally 0.5 ml) which varies according to the amount of sediment. Then a micropipette is used to take off 10 µl of the suspension and this suspension is then placed on two areas of 1 cm2 on a slide; there are two samples so there will be four smears altogether. The smears are dried at 40 C; when they are completely dry, it is not necessary to use the cover slide. Then the fungi imperfecti are counted under the microscope (300 x), using an eye-piece equipped with a reticle; 100 fields are counted for each of the four smears, starting from one side of the smear, going towards the centre, and obliquely proceeding to the opposite side. Pollen granules, fungus spores and algae are separately registered for every field area. After calculating the mean value of the 400 fields examined, the figure obtained is compared to the value obtained from 1 g or 10 g of honey. The choice, here, depends on microscopic field dimensions, the total sediment volume after the honey was dissolved and the initial quantity of honey.

Honey is divided into five classes, on the basis of the total number of fungi imperfecti. Class I, with less than 20,000 elements/10 g, includes unifloral honey from underrepresented pollen species. Most kinds of floral honey and of nectar and honeydew honey fall into class II (20,000-100,000 elements/10 g). Overrepresented honey and honeydew honey fall into class III (100,000-500,000 elements/10 g) and honey obtained by pressing falls into either Class IV (500,000-1,000,000 elements/10 g) or class V (> 1,000,000 elements/10 g).

Quantitative analysis results can be expressed by a simple formula: for example, "45/41/0.4 - II" means that in 10 g of a particular honey variety there are 45,000 pollen grains, 41,000 fungus elements and 400 algae; from this we can deduce that this honey variety is a mixture of nectar and honeydew and that it falls into class II. A nectar honey formula is "50/1/0 - II" and a honeydew honey formula is "30/90/0.6 - III" (Maurizio, 1955).

Demianowicz method

Sample preparation and evaluation of the results with this method do not vary substantially from the Maurizio method. There are only two small differences between the two methods: firstly, Demianowicz recommends that an area of 1 cm2 be delimited with Indian ink before placing the smear on the slide; this keeps the suspension from overflowing and guarantees that the area is exactly 1 cm2. Secondly, instead of counting 100 fields, the reading is performed on parallel transversal bands; the number of bands chosen depends on the quantity of sediment: 2 when the honey is very rich in pollen (from the III class upwards), 4 for normal honeys (II class) and 8 for those poor in pollen (I class).

In order to make the results of melissopalynological analysis more precise, Demianowicz undertook very painstaking research on 70 samples of experimental unifloral honeys obtained from small nuclei of bees kept in a controlled environment and thus compelled to gather pollen from one botanical species only. The samples to be analysed were prepared by placing a drop of honey directly onto the slide which was carefully weighed before and after the drop was positioned.

In this way it was possible to ascertain the number of pollen granules in 1 g of perfectly pure unifloral honey in all of the species studied. On the basis of the values obtained, Demianowicz divided the different kinds of unifloral honey into 18 classes; these classes were defined by means of conventional coefficients typical of each class (K = number of the pollen granules per gramme of honey) and ordered in a geometrical progression at the rate of 2. The very strongly underrepresented honeys fall into the first 3 classes (K values ranging from 112.5 to 450); normal honey varieties fall into the fourth (K = 900) and the fifth class (K = 1,800); slightly overrepresented honeys fall into the ninth class (K = 28,800) and the remaining classes include increasingly overrepresented honey varieties. The unclassified pollen types were assigned a coefficient K = 4,900, a purely statistical value.

By way of example, some of the botanical species studied by Demianowicz, together with their respective coefficients and classes, are listed below.

I112.5Cucumis sativus L., Robinia pseudacacia L.
II225Centaurea jacea L., Salvia nemorosa L., Tilia cordata Miller
III450Lamium album L., Ribes nigrum L., Salvia officinalis L.
IV900Allium cepa L., Centaurea cyanus L., Sinapis alba L.
V1,800Malus domestica Borkh., Onobrychis viciifolia Scop., Taraxacum officinale Weber, Trifolium repens L.
VI3,600Coriandrum sativum L., Echium vulgare L.
VII7,200Brassica napus L., Melilotus alba Medicus, Rubus idaeus Schott
VIII14,400Lythrum salicaria L.
IX28,800Lotus corniculatus L., Reseda lutea L., Reseda luteola L.
XIII460,000Cynoglossum officinale L.
XVIII14,745,600Myosotis sylvatica Hoffm.

In order to discover the real percentage of nectar in any given species in an apparently unifloral honey, the following formula is applied:

P(X) is the number of pollen granules of the X species in 1 g of the honey and K(X) is the coefficient typical of that species (Demianowicz, 1964).

Berner (1952) and Pritsch (1956-57) made several attempts to precisely quantify the real role of the different sources of nectar in the honey formation process. In fact, these Authors suggested experimentally determining a numerical rectification coefficient for every relevant honey-producing plant; this coefficient has to be introduced into a mathematical formula which takes into account all the pollens present in the honey being tested.

Even though it is theoretically exact, this formula is difficult to apply because the pollen contained in a given honey is subject to changes which make it impossible to adopt fixed characteristic values. This criticism applies to the Demianowicz classes, too; in fact, Demianowicz herself remarked that the pollen content of honey coming from the same botanical source could be subject to considerable changes, if conditions were modified. For example, the more numerous and active the colony, the higher the possibility of pollution; if the nectar source is far away from the beehive, the amount of pollen contained in nectar is reduced by the proventricule filtering action, etc.

Louveaux method

In order to adopt this method, Millipore filtering equipment and Millipore filters, with a diameter of 25 mm and 1 µ pores, microscope slides and cover glasses large enough to contain the filters, have to be available.

A given quantity of honey, chosen on the basis of the quantity of sediment (usually 10 g), is weighed, dissolved in 20 ml of distilled water, centrifuged and settled twice; the sediment is then dissolved in 10 ml of distilled water and placed in the filtering device, which is connected to an air-pump; the centrifuge tube is thoroughly rinsed with a small amount of distilled water; this water is then poured into the filtering machine and its internal walls are rinsed too. After filtering all the liquid, the filter is taken off and left to dry. A few drops of immersion oil are placed on a microscope slide and the filter is then placed on the slide where it becomes transparent; the compound is covered with a cover glass and the elements in 100 fields are counted; 800 x magnification is used so that the filter centre can be examined as well as its outer walls.

The number of fungi imperfecti found in the honey sample is given by the following formula:

N = total number of fungi imperfecti in the honey sample;
F = effective filter area;
n = total number of elements counted in all the fields;
f = microscopic field area;
a = number of fields examined (100).

The evaluation of the results is carried out in accordance with the Maurizio method (Louveaux et alii, 1978).


Reference materials

Palynologic analysis requires a good collection of reference materials. These include pollen samples taken from the different botanical species and the laboratory equipment needed for undertaking qualitative and quantitative analysis. In order to get the preliminary materials, it is advisable, though not always feasible, to collect un-opened flowers; anthesis must take place in a closed environment so that other pollens present in the air, especially those from anemophilous plants, do not contaminate the pollen samples. So either the ripe anthers, or the entire flower if very small, should be picked. There are two methods of sample preparation: the acetolytic one and the one prescribed by the International Commission for Plant-Bee Relationship (Louveaux et alii, 1978).

With the acetolytic method (flowers or anthers), first of all described by Erdtman (1952), the perfectly dry sample is placed into a centrifuge test-tube, and 4.5 ml of acetic anhydride and 0.5 ml of concentrated sulphuric acid are added; the compound is boiled in a bain-marie for about two minutes, centrifuged (5 minutes at 2,500 rpm), and then settled. Next the test-tube is filled up with distilled water plus two or three drops of acetone to prevent froth from forming on the surface and another centrifugation is carried out. The compound is settled, dissolved with 3 ml of a solution containing glycerine and distilled water and placed on a filter paper; the pollen is then picked up with a small fragment of glycerinated jelly and placed on a microscope slide and slightly heated with a flame until the gelatine melts; it is then covered with a cover glass and luted.

With this method the cytoplasm and the intina of the granules are destroyed; the exina is made darker by the acetolythic treatment and its structural and sculptural details can be seen clearly.

In the second method the sample is placed on a watch glass and washed with ether; after the remaining ether has evaporated, the pollen is picked up with a needle and a fragment of glycerinated gelatine and placed on a microscope slide where it is melted at 40 C; the compound is then covered with a cover glass and luted. The ether washing, used to eliminate fats, can be suppressed; in this case the granules retain a surface layer of oil which can give them colour nuances typical of their botanical species.

With this method it is possible to conserve the cytoplasm of the granules which sometimes presents useful typical features, but sporodermis details are not seen clearly.

In melissopalynology a magnification of 400 x is normally sufficient to discover the botanical source of any given honey; specialized analysis is only necessary in doubtful cases, i.e. when one needs to know more about pollens that are difficult to classify (1000 x).

In fresh honey the pollen resembles the results of the second method, while in a honey pollen that is several years old and no longer has its cytoplasm, the pollen resembles an acetolythic compound. It is, therefore, advisable that every sample collected be prepared in conformity with both the methods described above. After a certain time the pollen granules are subject to swelling and this alters their dimensions (the swelling is quite independent of the preparation method used). These compounds can still be used as reference materials, but it is advisable to keep a certain quantity of dried flowers of each species wrapped in paper, so as to be able to prepare further slides if they are needed. This herbarium should be arranged in alphabetical order to facilitate reference.

It is practically impossible to prepare a pollen collection which covers all botanical species. Pollen shapes, however, are relatively homogeneous within any given botanical family and so a pollen collection in which every family is represented, which contains the most widely diffused genera and species and the anomalous and typical pollen shapes, is perfectly adequate. If a researcher wants to study foreign honeys as well as domestic ones, the collection has to be increased: at least one sample of the better-known shapes, types, groups or genera of honey should be included in it.

In addition to the collection of the different kinds of pollen prepared according to the two methods described above and to the papers containing the dried flora samples, a melissopalynologist should have a collection of honey products and a collection of the different types of honey frozen in small containers.

A photographic archive could be useful, though it must be kept in mind that 4 1000 x photographs are needed for each pollen type and that in these photographs the exina sculpture, the polar view contour (equatorial optic section), the apocolpium, the equatorial view contour (meridian optic section) and the mesocolpium must be clearly seen.

All this material should be registered in a suitable card-holder to make reference easy (Ricciardelli D'Albore and Persano Oddo, 1978).


Quality honey with controlled origin

It is well-known that in most cases microscopic analysis of a honey sample makes it possible to identify the area where this honey was produced and its nectar sources with considerable accuracy.

Some years ago researchers realized that melissopalynology could be used not only to ascertain whether a honey sample was produced in Italy or elsewhere and whether it was unifloral or multifloral, but also to designate its geographical origin quite precisely and so the idea of quality honeys with controlled geographical denomination began to take shape (Ricciardelli D'Albore, 1987).

After a series of preliminary experiments it was realized that descriptions based on reliable data, i.e. a constant pollen spectrum, were essential if satisfactory results were to be obtained. The main factors determining spectrum variability are as follows:

  • human activity: even over a short time the phyto-sociological situation in a given area can be significantly modified; the impact is quite noticeable in the agro-ecosystem, but it is felt much less in first-class ecosystems. An example: sunflowers were introduced into central Italian agriculture about ten years ago, as a turn-over crop which soon became predominant but now they are disappearing. Flower association in mountain pastures, on the other hand, has practically remained unchanged;
  • vegetational dynamics: quite rapid in the agro-ecosystem, but very slow in first-class ecosystems except in cases of catastrophic events such as landslides, fires and floods;
  • the Rinchota population trend: it generally reaches its maximum density every 3-4 years, though of course it depends on the particular sucker insect species;
  • climatic conditions: condition the activity of the gatherers and the quantity of nectar secreted by the flora.

Experiments performed in this field indicate that about 3-4 years are required to describe a honey (only rarely is a longer period necessary). Furthermore, beekeepers have to observe the following directives:

  • honey has to be produced continuously by the same beekeepers in the same zone, using the same processing techniques unless, of course, the technical experts and analysts suggest they be improved. This is to ensure that honey obtained when this experimental phase ends is not only geographically characterized but also of excellent quality;
  • after the experimental phase is over, honey produced in any given area should be controlled at random by an analyst. This is also a way of checking the beekeepers' reliability; some of them may be tempted to increase the quantity of honey they can release on the market by obtaining honey produced elsewhere; this abuse of a geographically denominated trademark would severely damage the interests of the other members of the association.

Several beekeepers' associations have shown a heartening interest in this problem and have considered adopting a trademark to certify the quality and the geographical origin of the honey produced by their members. These associations were also of the opinion that these trademarks could be a means for promoting Italian honeys, especially the multifloral varieties, which are not easily characterized by physical, chemical and organoleptic parameters.

Some honeys have already been geographically characterized: acacia and chestnut honeys from the Varese province, honeys from the Asiago plateau, from the Veneto region, from the Gubbio - Gualdo Tadino area (Umbria); chestnut, acacia and multifloral honey from the Lunigiana area, etc. In some cases the experimental phase is still in progress (Ricciardelli D'Albore, 1989, 1991) (Figures 10-13).

Where the experimental phase has been concluded the results have met with the beekeepers' full approval; a certain number of associations have increased their profits considerably. Obviously they avail themselves of the analyst's certificates which attest the quality and origin of the honey and guarantee the authenticity of the description provided on the label of honey put on the market.

It should be noted that the beekeepers' associations favouring innovations only seem to operate in northern and central Italy; the ones in southern Italy do not seem to be interested. The reason for this may well lie in the fact that the conventions organized to explain and promote this experimental project have all been held in central and northern Italy.

The production of high quality, geographically characterized honeys is definitely a good way of promoting the different varieties of Italian honey and is in accord with the recent EEC directive promoting quality food production (Reg. no. 1082, 1992).


Figure 10. Pollen spectrum of the controlled geographical origin honey: different kinds of Castanea sativa Miller honey coming from the province of Varese.

Figure 11. Pollen spectrum of the controlled geographical origin honey: different kinds of Robinia pseudacacia L. pollen from the province of Varese.

Figure 12. Pollen spectrum of the controlled geographical origin honey: different kinds of honey from the Gubbio and Gualdo Tadino district (Umbria).

Figure 13. Pollen spectrum of the controlled geographical origin honey: different kinds of honey from the Asiago plateau (Vicenza).


Mediterranean melissopalynology

Many researchers on honeys of the Mediterranean Basin have provided important contributions to the knowledge of the wild and cultivated bee forage (Accorti et al. 1986; Barbier 1958; Battaglini and Ricciardelli D'Albore 1967, 1970a, b, 1971a, b, c, 1972, 1981; Battaglini et al. 1973; Belmonte et al. 1986 a, b; Bermudez 1978; Bolchi Serini and Salvi 1987; Borque 1982; Damblon 1988; De Leonardis et al. 1982, 1984 a, b, 1986, 1988, 1989 a, b; Espada 1984, 1986; Ferrazzi 1974a, b, 1977, 1982, 1983, 1986; Ferrazzi and Manino 1977; Ferrazzi and Marletto 1985; Ferrazzi et al. 1978; Ferrazzi and Patetta 1979; Ferrazzi, Patetta and Manino, 1990; Ferrazzi and Priore 1987; Floris et al. 1991; Forlani 1981; Genier 1966; Gomez Ferreras 1985 a, b, 1988; Gomez Ferreras and Saenz 1980, 1985; Huibobro and Simal 1984 a, b; Ibrahim Sabry 1976; Intoppa et al. 1976; Longhitano et al. 1982, 1986 a, b; Louveaux and Abed 1984; Louveaux et al. 1978; Louveaux and Vergeron 1964; Maurizio 1960, 1968, 1984; Maurizio and Louveaux 1960, 1961, 1962, 1963, 1964, 1965; Ortiz 1985; Oustuani 1966; Perez de Zabalza 1988; Perez de Zabalza and Gomez Ferreras 1988; Perez de Zabalza and Ricciardelli D'Albore 1990; Persano Oddo and Ricciardelli D'Albore 1974, 1975, 1987, 1989; Pozo Lora 1970; Priore and Ricciardelli D'Albore 1989; Reille 1992; Ricciardelli D'Albore 1974, 1980, 1983, 1985, 1989a, b, 1991, 1997; Ricciardelli D'Albore and Persano Oddo 1978; Ricciardelli D'Albore and Piastrelli 1977; Ricciardelli D'Albore and Priore 1980, 1984; Ricciardelli D'Albore and Quaranta 1991, 1992; Ricciardelli D'Albore and Tonini D'Ambrosio 1973; Ricciardelli D'Albore and Vorwohl 1980; Sala Llinares 1988; Sanchez Sanchez 1982; Serra et al. 1986; Vergeron 1964; Vieitez 1950;Vorwohl 1973 a, b, 1981; Zizza et al. 1985).

In 1980 a work on the well known Mediterranean unifloral honeys was published, considering a territory band of about 250 Km far from the coast toward the inland of the countries facing the Basin (the band was wider or narrower depending on the influence of the Mediterranean climate) (Ricciardelli D'Albore and Vorwohl 1980). The new contributions have been published prevalently in Italy and Spain. Detailed morphopalynological information are available in few publications.

This work aims to describe the most frequent fresh pollens found in the Mediterranean honeys; furthermore one will make a sign of the pollen combinations (pollen spectra) (see also References); it is as well in plan to give a contribution to the knowledge of the pollen gathered by the honeybees.

More than two hundred pollens characteristic of the Mediterranean unifloral and multifloral honeys have been selected. They are of typical Mediterranean species or of cultivated and ornamental plants long before imported, which now can be considered integrated in the Mediterranean vegetation (Polunin and Huxley 1968). One will also give some phytogeographical information on the Mediterranean countries.

In this work about melissopalynology each pollen has been illustrated prevalently in two positions and four focus with a M.O. at 1000 x: polar and equatorial view, equatorial and meridian section. Sometimes particularities of the sculpture and aperture and illustrations at 400 x or 1000 x are given, too. Rarely the polar view is lacking, because practically in this position pollen is sometimes difficult to be found (see Umbelliferae). The slides belong to the collection of the Agricultural Entomology Institute, Apiculture Section-University of Perugia (Italy), and prepared like the well known method for fresh pollens by Louveaux et al. (1978).
All slides are approximately six weeks old; they have been coloured with basic fucsine (without colour the particularities of the aperture and the sculpture are less clear).
Unfortunately each pollen has its own capability of colour absorption; therefore the intensity of the colour is variable; so some illustrations are more or less dark, but yet the understanding and the interpretation of the details of each pollen grain are equally assured. For each pollen 50 measures have been made for P and E min., max and average (Punt et al. 1994).

Phytogeography of the Mediterranean countries

Because of many microclimates, so that the deciduous forest may involve both the coast and the typical Mediterranean flora and the inside valleys, the parameter adopted in delimiting the typical Mediterranean zones (250 Km) reaches only an approximate value (Ricciardelli D'Albore and Vorwohl 1980).

The Mediterranean Basin is characterized by a remarkable wealth of vegetation, prevalently wild and partially represented by imported plants for cultivation and for ornament. Generally it is allowed to say that where isolated items of Olea europaea L., Quercus ilex L., Quercus coccifera L. and Pinus halepensis Miller live (or coexist) there the influence of the Mediterranean climate is determining. As a consequence of phenomena as evolution or involution and the migrations of the species through the millenary, also with the anthropic influence (cultivation) today the following vegetable communities are present: forest, maquis, "gariga", steppe, cultivated fields and ornamental plants associations (Polunin and Huxley 1968).

Evergreen forest (also considered primary maquis)
Tree: Quercus ilex L. (Circummediterranean), Q. coccifera L. (Eastern), Q. suber L. (West and Central Mediterranean), Pinus halepensis Miller (hot zones) are found.
Underbrush: Arbutus unedo L., Phyllirea spp., Rhamnus alaternus L., Viburnum tinus L., Clematis vitalba L., Lonicera spp., Smilax aspera L., Tamus communis L., Juniperus spp., Pistacia spp., Rosmarinus officinalis L., Lavandula stoechas L. Locally are also present primary maquies of Pinus spp., Erica spp., Cistus spp. and Laurus nobilis L. (Polunin and Huxley 1968).

Maquis. This is a secondary vegetable community distinguishable as: maquis with tree and shrubs (Arbutus spp., Cercis siliquastrum L., Olea spp., Pinus spp., Myrtus communis L., Erica spp., Genista spp., etc.); maquis without high vegetation (Rosmarinus spp., Phlomis spp., Paliurus australis Gaertner, Cistus spp.); maquis typically with many Cistus species; finally heterogeneous maquis (many species with Crataegus monogyna Jacq.) (Polunin and Huxley 1968).

"Gariga". So are called the dry and arid wide Mediterranean areas, where extremely rustic species live, also spread in stony and rocky zones. There can live hundred of species, especially aromatic plants such as Thymus spp., Salvia spp., Satureja hortensis L. and S. montana L., Hyssopus officinalis L., Allium spp., Ruta graveolens L., many Liliaceae, Orchidaceae, Euphorbiaceae, Leguminosae (prevalently shrubs) such as Calycotome spp., Ulex europaeus L., etc. (Polunin and Huxley 1968).

Steppe. It is the anthropised area (effect of the man and the cattle); after soils denudation a new vegetables association forms again. There are in this area many Compositae, Gramineae, Cruciferae and Umbelliferae (above all we remember Carduus spp., Centaurea spp., Cirsium spp., Ferula spp. and Verbascum spp.; finally, in the most lowered grounds, Urginea maritima (L.) Baker and Asphodelus spp. (Polunin and Huxley, 1968).

Cultivated and ornamental plants. They are many and often not endemic, but now to be considered such as Mediterranean flora. Above all we remember the orchards (Citrus spp., Eriobotrya japonica (Thunb.) Lindley, Malus domestica Borkh., Pinus spp. and Prunus spp.); recently also Musa x paradisiaca L., Mangifera indica L., Persea spp., etc., besides many species destined to the human feeding and cultivated forages for cattle. Some ornamental plants that can be considered "Mediterranean" are Eucalyptus spp., Acacia spp., Palmae, Agave spp., Opuntia ficus-indica L., Schinus molle L., Bouganvillea spp., Robinia pseudacacia L., Melia azedarach L., etc. (Polunin and Huxley 1968). These cultivated wild species and many others in the Mediterranean areas represent bee forage, on which honeybees gather nectar and/or pollen and/or honeydew and/or propolis (Ricciardelli D'Albore and Persano Oddo 1978).

Pollen presenting problems in melissopalynology

There are many factors influencing the under/overrepresentativity of pollen in honey. Apart from the different kinds of pollution, universally recognized, it could be useful to describe these other factors in detail for both readers and relatively inexperienced analysts.

The above mentioned factors can interact, either high-lighting or neutralising a given phenomenon (Ricciardelli D'Albore, 1983).

Large size pollens

(Abutilon, Acacia, Agave, Albizia, Althaea, Arbutus, Arctium, Asphodelus, Calystegia, Carduus, Carthamus, Cephalaria, Cirsium, Citrullus, Cornus sanguinea, Cucumis, Cucurbita, Cynara, Echinops, Epilobium, Hibiscus, Geranium, Gossypium, Knautia, Lavatera, Linum, Liriodendron, Lonicera, Malva, Musa, Ocimum, Oenothera, Opuntia, Scabiosa, Sesamum, Silybum, etc.). The nectars are poor in pollen and bees tend to filter them through their Zander valve; this phenomenon is called underrepresentativity.

Practical problems: bearing in mind that many of the genera listed above are not widespread in Italy and that they flower in periods when bees do not produce honey, there may be some problems with Asphodelus, Cephalaria, S-shaped Compositae in general, Epilobium, Liriodendron and Opuntia. Organoleptic and physical-chemical analysis have to be carried out with great care on Carduus and Asphodelus honeys, which are produced in substantial quantities, especially in Sardinia. The same comments apply to pollen coming from abroad, even though very little is known about it. Musa honey, in particular, always has a percentage of pollen lower than 2%; sometimes there is no pollen at all so different analytical methods, such as sensorial analysis, are used to identify Musa honey.

Very small size pollens

(Alkanna, Amorpha, Asparagus, Castanea, Cynoglossum, Echium, Eucalyptus, Galega, Lotus, Smilax, Sophora, Tamarix, etc.). Nectar is incredibly rich in pollen and bees filter it only partially; this phenomenon is commonly called pollen overrepresentativity. Practical problems: Amorpha and Galega nectar secretions need to be further studied; other phenomena are mostly known.

Insufficient or absent or aborted pollen

(Citrus, Lavandula, Pittosporum, Robinia, Rosmarinus, Salvia, etc.). Underrepresentativity is accentuated. Practical problems: Pittosporum and Salvia are not very widespread; Citrus, Lavandula and Robinia require all available analysis because there are some citrus fruits honeys with considerable amounts of Citrus pollen (though these cultivars are in extinction), while most of them have a very low percentage of Citrus pollen. This phenomenon can be also observed in Lavandula and Robinia honey; the organoleptic analysis have to be carried out very carefully; excessively aromatic citrus fruit and lavender honeys should be avoided because they have probably been subjected to fraudulent practices; Lavandula honey in particular requires an attentive analysis; Rosmarinus does not create particular problems.

Monoecious and dioecious plants

(Asparagus, Bryonia, Citrullus, Cucumis, Cucurbita, Ecballium, Palmae, Salix, etc.). Bearing in mind that bees also alight on flowers without pollen (female flowers), the pollen representativity is obviously reduced. Practical problems: the Palmae nectar secretions need to be further studied as do the representativity of Salix pollen. Other genera are not widespread.

Extrafloral nectar

(Prunus laurocerasus, Gossypium, etc.). Underrepresentativity taken for granted.
Practical problems: very little is known about this phenomenon in the non endemic species.

Particular flower morphology, unfavourable for pollen pollution

(Abutilon, Arbutus, Asphodelus, Borago, Datura, Digitalis, Epilobium, Lonicera, Oenothera, Paulownia, etc.). Different types of more or less covered nectars (Asphodelus, Epilobium), upside-down flowers (Abutilon, Arbutus, Borago), and very large corollas (Datura, Digitalis, Oenothera, Paulownia) reduce pollen representativity.

Arbutus and Asphodelus present practical problems: quantitative pollen analysis requires particular care because it is connected to the organoleptic and physico-chemical analysis. The other genera are not widespread.

Special gathering situations

A. (Cruciferae, Medicago, Oenotheraceae, etc.). Bees learn to gather close to and on the sides of the flower corollae, thereby avoiding direct contact with the pollen.

Practical problems: they arise only for lucerne and sometimes for Cruciferae whose cultivation is spreading everywhere. All the available analysis are necessary.

B. (Agrostemma, Cestrum, Chamaecytisus, Dianthus, Narcissus, Nicotiana, Onosma, Saponaria, Vicia faba). These genera can be visited by bees only on the condition that bumble bees have previously made holes in the corolla base (the so-called "robber" bumble bees) (Ricciardelli D'Albore, 1983). Practical problems: only for Vicia faba and other Vicia species in areas where these cultivars are widespread. The other genera are not particularly widespread.

Dominant pollens of nectarless species

In Europe honeys with dominant pollens are quite common (Cistus, Hypecoum, Gramineae, Quercus, Rumex, etc.). A good knowledge of nectar secretion makes it possible to avoid wrong diagnoses. The case of Cistus is quite controversial because it would seem that nectaripherous species do exist (Cistus monspeliensis, etc.).

Beekeeping techniques

(Excessive and prolonged feeding with nectar syrup, honey-extraction from honeycombs in the hive, special breeding situations where there is no division between the hive and the honeycomb, beehive honey-extraction, etc.).

In the majority of cases there is an anomalous pollen enrichment of the honey and high percentages of pollen from nectarless plants, gathered by the bees to satisfy their protein requirements.

Practical problems: organoleptic analysis, quantitative pollen analysis and the ash content are very important for identifying excessive feeding with nectar syrups; organoleptic analysis makes it possible to overcome difficulties inherent in the interpretation of an altered pollen spectrum. In Italy this phenomenon is restricted to the south and the islands; it is particularly widespread in Sardinia. The extraction of honey from the nestcombs is very common in many Mediterranean areas. We have also to remember a typical beekeeping technology in Spain, whose honeys are always overcontaminated by pollen, frequently of nectarless species.


(Addition of saccharose or isomerose, artificial honey, addition of metals, dyes, pollen, flavours, pH alterations, honey with molasses, etc.).

Practical problems: in most cases organoleptic, microscopical and physical-chemical analysis, when carried out together, enable the fraud to be detected. In particular cases (sugars), sophisticated devices, mass spectrometer can be useful. Abroad, the frauds perpetrated in the Middle East (Syria) are well known (Ricciardelli D'Albore 1997, modified).

Filtered honey

This technique is widely used in the USA; it renders melissopalynological analysis absolutely impossible because the filters used (Diatomee soil, special sands, etc.) retain 100% of the pollen granules (Ricciardelli D'Albore and Persano Oddo, 1978).


Main Mediterranean unifloral honeys and their pollen spectra

A contribution to the knowledge of the Mediterranean unifloral honeys and their pollen spectra has been all ready partially provided (Ricciardelli D'Albore 1997, Ricciardelli D'Albore and Vorwohl 1980). The more or less frequent pollens found are as follows: *

(*Plants are listed prevalently for genus; they are listed for species if they vegetate as a single species or if they have to be distinguished from other less or not important species living in a country.)

Delimitation of Mediterranean climate


Spain (1) Anthyllis lotoides, Citrus, Diplotaxis, Echium, Erica umbellata, Eucalyptus, Genista, honeydews, Lavandula, Onobrychis viciifolia, Pimpinella, Prunus, Rosmarinus officinalis, Rubus and Thymus.

Other frequent pollens: Asparagus, Brassica, Cistus, Erica multiflora, Helianthus annuus, Hypecoum, Lotus, Olea europaea, Plantago, Quercus, Salix and Ulex.

France (2) Brassica, Erica arborea, honeydews, Lavandula, Onobrychis viciifolia, Rosmarinus officinalis and Trifolium repens.

Other frequent pollens: Centaurea jacea, Cistus, Cornus sanguinea, Diplotaxis, Fraxinus, Genista, Lotus, Malus domestica, Melilotus, Mercurialis, Quercus, Rubus, Salvia, Trifolium pratense and Ulmus.

Italy (3) Common honeys: Arbutus unedo, Asphodelus microcarpus, Carduus, Castanea sativa, Citrus, Eucalyptus camaldulensis, Eucalyptus viminalis, Hedysarum coronarium, Helianthus annuus, honeydews, Medicago sativa, Onobrychis viciifolia, Robinia pseudacacia, Taraxacum officinale, Thymus, Trifolium incarnatum and Trifolium repens.

Rare honeys: Arctium minus, Asparagus acutifolius, Brassica, Centaurea solstitialis, Ceratonia siliqua, Cichorium intybus, Cirsium arvense, Daucus, Diplotaxis, Echium, Erica arborea, Eriobotrya japonica, Galactites tomentosa, Hedera helix, Inula viscosa, Lavandula, Lotus, Malus domestica, Myrtus communis, Odontites, Paliurus, Prunus, Rosmarinus officinalis, Rubus, Salix, Satureja, Sideritis, Solidago virgaurea, Stachys, Trifolium alexandrinum, Trifolium pratense, Viburnum and Vicia.

Other frequent pollens: Aesculus hippocastanum, Ailanthus altissima, Allium, Centaurea jacea, Centaurea cyanus, Chamaerops, Cistus, Clematis, Cornus sanguinea, Crataegus monogyna, Cupressus, Ferula, Fraxinus, Galega officinalis, Ligustrum, Melilotus, Olea europaea, Rhamnus, Quercus, Ranunculus, Sambucus, Smilax aspera, Verbascum, Vitis vinifera and Zea mays.

Ex-Yugoslavia (Croatia and Bosnia) (4) Castanea sativa, Erica arborea, Fagopyrum esculentum, honeydews, Robinia pseudacacia, Rubus and Salvia officinalis.

Other frequent pollens: Carduus, Centaurea cyanus, Centaurea jacea, Cistus, Diplotaxis, Loranthus europaeus, Lotus, Olea europaea, Sedum, Tilia, Thymus, Trifolium pratense, Trifolium repens and Vitis vinifera.

Albania (5) Arbutus unedo, Astragalus, Castanea sativa, Erica arborea, Eryngium, Helianthus annuus, honeydews, Ononis, Rubus and Thymus.

Other frequent pollens: Brassica, Centaurea cyanus, Centaurea jacea, Echium, Hypericum, Lotus, Myrtus communis, Senecio, Trifolium pratense, Trifolium repens and Xanthium.

Greece (6) Carthamus, Castanea sativa, Citrus, Erica manipuliflora, Helianthus annuus, honeydews, Lotus, Sesamum and Thymus capitatus.

Other frequent pollens: Asparagus, Carduus, Cistus, Ephedra, Eucalyptus, Erica arborea, Hypericum, Olea europaea, Papaver, Polygonum dumetorum, Rhamnus, Tribulus terrestris, Trifolium alexandrinum and Trifolium repens.

Turkey (7) Castanea sativa, Centaurea cyanus, Ceratonia siliqua, Citrus, Erica manipuliflora, honeydews, Leopoldia, Lotus, Melilotus alba, Punica granatum and Sophora japonica.

Other frequent pollens: Carthamus, Cistus, Cytisus, Centaurea calcitrapa, Echium, Euphorbia, Galega officinalis, Myrtus communis, Petroselinum, Ranunculus, Raphanus, Smilax aspera, Thymus, Trifolium repens and Xanthium.

Cyprus (8) Citrus, Lavandula, Myrtus communis, Thymus capitatus and Vicia.

Other frequent pollens: Carthamus, Echium, Eucalyptus, Hypericum, Mercurialis, Papaver, Petroselinum, Salvia and Sinapis.

Syria (9) Petroselinum and Trifolium repens.

Other frequent pollens: Carthamus, Centaurea jacea, Euphorbia, Myrtus communis, Olea europaea, Pimpinella, Ranunculus, Salix, Sinapis, Sophora japonica, Trifolium pratense, Vitis vinifera and Xanthium.

Lebanon (10) Citrus, Eucalyptus, Euphorbia, Salix and Trifolium repens.

Other frequent pollens: Astragalus, Astrantia, Carlina, Cistus, Malus domestica, Myrtus communis, Prunus, Ranunculus, Sinapis, Thymus and Trifolium pratense.

Jordan (11) Brassica and Eucalyptus.

Other frequent pollens: Artedia, Centaurea jacea, Citrus, Cucumis, Echium, Myrtus communis, Pulmonaria, Trifolium repens and Sinapis.

Israel (12) Citrus, Eucalyptus, Helianthus annuus, Malus domestica, Persea, Trifolium alexandrinum and Trifolium repens.

Other frequent pollens: Acacia, Allium, Arbutus andrachne, Asparagus, Centaurea solstitialis, Ceratonia siliqua, Foeniculum, Hypecoum, Inula, Linum, Medicago, Melilotus, Morus, Myrtus communis, Olea europaea, Phoenix dactylifera, Salvia, Trifolium pratense Gr. and Vitis vinifera.

Egypt (13) Citrus, Eucalyptus, Gossypium, Musa x paradisiaca, Phoenix dactylifera and Trifolium alexandrinum.

Other frequent pollens: Acacia, Bombax, Carthamus, Casuarina, Cistus, Cucumis, Diplotaxis, Echium, Euphorbia, Hypericum, Linum, Luffa, Mercurialis, Papaver, Punica granatum, Sesamum, Trifolium repens, Vicia and Vitis vinifera.

Libya (14) Acacia, Agave, Astragalus, Citrus, Eucalyptus, Hedysarum coronarium and Ononis.

Other frequent pollens: Caesalpinia, Cakile, Carduus, Carthamus, Diplotaxis, Echinops, Medicago, Musa x paradisiaca, Phoenix dactylifera, Schinus, Trifolium pratense and Trifolium repens.

Tunisia (15) Citrus, Cruciferae, Erica multiflora, Eryngium, Eucalyptus, Hedysarum coronarium, honeydews, Lavandula, Pimpinella, Prunus and Rosmarinus officinalis.

Other frequent pollens: Acacia, Daucus, Diplotaxis, Echium, Lotus, Olea europaea, Oxalis, Reseda and Vicia.

Algeria (16) Citrus, Eucalyptus, Rosmarinus officinalis and Trifolium repens.

Other frequent pollens: Acacia, Asphodelus, Ceratonia siliqua, Cerinthe, Chamaerops, Cistus, Diplotaxis, Echium, Hedysarum coronarium, Inula, Malus domestica, Pimpinella, Quercus, Rubus, Salix and Vicia.

Morocco (17) Citrus, Echium, Eryngium, Eucalyptus, Lavandula, Peganum harmala, Sapium and Thymus.

Other frequent pollens: Acacia, Arctium, Brassica, Cistus, Ephedra, Olea europaea, Pimpinella, Plantago, Quercus, Tamarix, Trifolium repens and Urginea.


Honeys coming from the Mediterranean countries may be different, according to the following presumptions:
  1. the pollen spectrum of a honey is completely or partially different as regards to one of another country;
  2. a honey contains pollens common to other honeys, but also others, belonging to species absent in the first country;
  3. the combinations and percentages of pollens in a honey are different as regards to another honey;
  4. in a honey, as regards to another, one or some pollens are lacking;
  5. a unifloral honey is exclusive of a limited country (Battaglini and Ricciardelli D'Albore 1973).

So the pollen spectra (unifloral or multifloral honeys) of a country are never equal to those of honeys coming from another one (see the above listed pollens of each country and see also references); the same considerations are possible for the aspects 2, 3 and 4.

As regards to the unifloral honey (5) we can consider that Anthyllis lotoides and Erica umbellata honeys are present only in Spain; Asphodelus, Taraxacum, Cichorium, Eriobotrya, Satureja etc. only in Italy; Salvia only in ex-Yugoslavia (Croatia and Bosnia), such as Sesamum in Greece, Punica in Turkey, Euphorbia in Lebanon, Persea in Israel, Gossypium and Musa in Egypt, Agave in Libya, Erica multiflora in Tunisia, Peganum in Morocco, etc.

In conclusion although the differentiation of some honeys of contiguous countries may be difficult, in many cases the differentiation among the honeys is possible; but it requires a good knowledge in melissopalynology and a valuable collection of pollens and honeys slides. Other information on this topic are given in a recent work (Ricciardelli D'Albore 1997).


Mediterranean unifloral honeys: distribution maps
Mediterranean honeydews

Honeydews are excretions made by piercing and suction insects (Rinchota Homoptera), which use their piercing oral apparatus to attack plants and suck out the phloem sap which is rich in nourishing substances, especially aminoacids. To satisfy their protein needs, these insects are compelled to suck large amounts of this phlohema sap which contains only 1-2% of proteins and is rich in water and sugars. Some Rinchota (for example Coccides) have a complex digestive tract, with a filtering chamber which connects the ingluvies to the rectum. Most of their food passes across this filter and, as a consequence, their excretions become richer in water and saccharose (the carrier sugar in plant sap).

In most other piercing and suction insects (for example Aphids) there is no filtering chamber in the digestive tract; as a result, their food passes through the entire abdominal duct, which digests, absorbs and transforms most of the sugars. The chemical composition of the honeydew produced by these insects differs considerably from that of the insects whose digestive tract has a filtering chamber. Many other factors, related to insect and plant biology, affect the quantity and the quality of the honeydew produced. Microscopic fungi and algae develop on these nectar excretions. These fungi imperfecti can then be traced in the honeydew sediment.

One often hears of the damages that piercing and suction insects inflict on plants are less than is commonly thought. Even though an insect attack may be harmful (for example fumagines can develop), it is equally true that plants recover quite well by themselves (in agriculture the situation is different, because production quantity and quality have to be taken into account). Moreover, in a forestry ecosystem hundreds of useful insect species feed on honeydews, which consequently play an important role in the food chain.

Honeydew production depends directly on the population trends of these insects and these vary greatly from one species to another. A large quantity of honeydew can only be produced when the insect population density is very high; it should be noted that even when the insect population is at its maximum, very adverse meteorological conditions (such as strong storms) can totally ruin honeydew production. It is not advisable to practise beekeeping in a forest if there is no possibility of obtaining a prognosis of the honeydews likely to be produced in the immediate future. Experts know what techniques to use to obtain a reliable prognosis. Basic entomological knowledge is required in this field: insects in every development phase can be quantified per surface unit (for instance a 1 meter branch, etc.) or by the number of drops of honeydew which fall onto a plastic sheet per surface unit and per time unit. Prognosis therefore requires meticulous observation in the field and enables a beekeeper to estimate how many bee colonies can be transferred into any given area, for how long they should remain there and what production levels can be expected.

Some honeydews are produced in large quantities in a short time (conifers or Metcalfa honeydews), but one generally obtains an abundant honeydew production every four years (many broadleaves). Sometimes the Rinchota population explosion happens over very long periods of time (6-8 years, Chestnut honeydew); however, since Rinchota trends depend on numerous biotic and abiotic factors, no fixed guide-lines can be laid down for any of the species, so a reliable prognosis can only predict whether there will be a large production of honeydew in a given year (Ricciardelli D'Albore 1997, modified).


The main types of honeydew

This paragraph provides a description of honeydews that are more or less constantly produced in large quantities and which contemporaneously attract bees.

  • Silver fir (Abies alba Miller) honeydew.
    It is considered one of the best honeydew varieties in Mediterranean zones; the insects responsible for its production belong to the Cinara genus (Cinara pectinatae Nordlinger plays a particularly important role). In Italy this honeydew variety comes from the Tusco-Emilian Apennines area. Silver fir honeydew honey is almost always contaminated with chestnut nectar.
    Only very rarely does melissopalynological analysis result in a PK/fungus element ratio > 1 in the honey sediment. Mention should also be made of the so-called silver fir honeydew (A. bornmuelleriana) produced in Greece and Turkey.
  • Oak-tree (Quercus spp.) honeydew.
    It is obtained periodically in good quantities. Its quality is inferior to spruce and silver fir honeydew from the organoleptic point of view. Honey made with oak-tree honeydew does not remain liquid for long; it crystallizes firmly, forming large crystals. Oak-tree honeydew is always very poor in fungi substances, because bees usually gather this sugar source for a short period of time and when the honeydew has just been secreted and is, therefore, quite pure. The insects which play the largest role in oak-tree honeydew production are Tuberculatus annulatus Hartig. and T. borealis Krzywiecz. Holm-oak (Quercus ilex L.) honeydew is probably the worst of the oak honeydews. Oak-tree honeydew is quite common in central and southern Italy, in Spain and other Mediterranean countries.
  • Metcalfa pruinosa (Say) honeydew.
    It is one of the few honeydews whose name comes from the insect which produces it. This insect was introduced into Italy recently and quite by accident. It is polyphagous, i.e. it feeds on many spontaneous plants and also on plants cultivated for agricultural and ornamental purposes. Generally honey made from this honeydew is of poor quality (it has a typical molasses flavour); if, however, these insects also attack lime or maple-trees, its quality can improve. M. pruinosa is widespread in Veneto and is quickly spreading to other regions.
    The melissopalynological characteristics of Metcalfa honeydew are the same as those for spruce honeydew.
  • Chestnut (Castanea sativa Miller) honeydew.
    Given the diffusion of this species in Italy, one would expect a high production of Chestnut honeydew honey. This does not happen for two reasons. Firstly, bees are not greatly attracted by this glucidic source, and secondly high production rates can only be reached after a long period of time. The most important piercing and suction insect for chestnut trees is Myzocallis castanicola Baker; this insect only reaches a high population density after quite a few years.
  • Willow-tree (Salix spp.) honeydew.
    It is excellent, but not widespread in Italy (there is small-scale production in Piedmont, Tuscany and Umbria). The honey which derives from willow-tree honeydew is very rich in fungi imperfecti and has a typical liquorice aftertaste. In ex-Yugoslavia this kind of honey is produced in very substantial quantities. The piercing and suction insect is Tuberolachnus salignus Gmelin.
  • Pine (Pinus brutia Ten.) honeydew.
    Despite the fact that there are many species of Pine in Italy (including Aleppo), and that specific piercing and suction insects belonging to the Cinara genus are present, there is no evidence that any Pine honeydew honey is actually produced. In Greece, however, large amounts of this type of honey are obtained from Aleppo pine, thanks to the activity of piercing and suction insects belonging to the Marchalina hellenica Gennadius species.
    Pine honeydew honey contains an enormous number of hyphae and spores; the hyphae are particularly large and belong to a fungi species usually not found in the various kinds of honeydew honey produced in Italy (a marker element).
  • Wheat (Triticum spp.) honeydew.
    Cereals are generally well controlled by man but under certain conditions they can be strongly attacked by piercing and suction insects, such as Sitobion avenae F. (very rarely though). When this does happen bees gather a very tasty honeydew, while farmers have to evaluate whether they should adopt expensive treatments or not; usually the resulting wheat weight loss is not such as to justify these treatments because they result uneconomical. Cereal honeydew honey is clearer than the better-known honeydew honeys (it looks like spruce honeydew honey) and contains many fungi imperfecti.
  • Citrus fruits (Citrus spp.) honeydew.
    These cultures are also controlled by farmers. However, a very tasty honeydew is produced by various Rinchota (such as Aleurothrixus, Planococcus, Yceria, etc.) in abandoned Citrus groves.
    About ten years ago researchers found some Citrus fruit honeydew honey which came from Calabria and Sicily and from Tunisia.
    This type of honey is also quite clear (Ricciardelli D'Albore 1997 modified).


Occasional honeydews

This paragraph deals with honeydews that do not strongly attract bees, even though produced in large amounts, and with others that attract bees but do not give rise to a unifloral honey, because the vegetative species is relatively scarse in a given area.

  • Maple tree (Acer spp.) honeydew.
    Only Acer pseudoplatanus L. honeydew is of some interest to bees. The piercing and suction insects responsible for its production are Peryphyllus acericola Walker.
  • Judas tree (Cercis siliquastrum L.) honeydew.
    Produced by Psylla pulchella Löw, this kind of honeydew greatly attracts bees, especially in parks and in certain central Italian valleys where it forms spontaneously (for example Valle del Nera in Umbria). It should be noted that this is one of the relatively infrequent cases where the piercing and suction insect not only attacks the lower blade of the leaf (typical in the broadleaves), but also the upper one and it then attacks the siliquae produced by flower fecundation.
  • Fig tree (Ficus carica L.) honeydew.
    The piercing and suction insect Homotoma ficus L. produces an abundant honeydew which attracts bees, but it is of only relative and sporadic interest in Central-southern and Insular Italy and in East Mediterranean.
  • Poplar (Populus spp.) honeydew.
    Bees visit poplar-woods in spring to gather pollen and in summer to collect propolis; they are not attracted by the honeydew produced by piercing and suction insects belonging to the Chaitophorus genus (C. populeti Pamper and C. tremulae Kock).
Mediterranean honeydews honeys: distribution map


Botanical and geographical origin of pollen

Pollen is the only proteic food within the beehive; as a consequence, it plays a fundamental role in feeding the colony, whose biology is entirely conditioned by this factor. In fact, pollen is used for feeding the larvae and the young bees. It contributes to body growth in general and is a determining factor in the development and the functionality of certain organs such as the adipose body, ovaries and in particular the hypopharingeal glands; these glands play an important role in royal jelly secretion; royal jelly is used for feeding the larvae for the first three days of their life and provides the queen bee with nourishment for the entire larval and imaginal phase.

Special devices can be set up in the beehive to enable a careful analysis, both quantitative and qualitative, of the pollen collected. These devices, known as "pollen traps", can be of two different types, but all are based on the same mechanism: the beehive entrance is closed with a grill whose small holes are round or star-shaped; they are dimensioned in such a way that the bees can pass through, but some of the small balls of pollen they are carrying fall off into a little box situated below the entrance to the hive. The most commonly used trap today is collocated below the floor of the beehive; with this trap it is possible to gather good quantities of quite dry pollen that is practically parasite free (moth).

In the period immediately after the grill is installed, almost all the pollen gathered ends up in the trap, but after some days, the gatherers learn how to enter the hive without losing their entire pollen load. So the trap yield settles at about 10%; this means the trap can be used, at least for a certain period of time, without substantially damaging the colony, especially as the colony makes up for its losses by working more intensely.

The contents of the trap are removed at regular intervals and carefully analysed. Before being weighed they are dried in an oven for a certain time to avoid the weight being affected by the humidity content. The small pollen loads are then divided into groups on the basis of their colour, shape and consistence and for each of these groups a specimen is prepared for microscopical analysis. Some small balls of pollen are dissolved in a few drops of distilled water on a slide. Part of this solution is poured into a Pasteur pipette, dried on another slide and prepared with glycerinated jelly. The slide is then covered with a cover slide and luted with wax or enamel. Many studies have been carried out on pollen gathering (Ricciardelli D'Albore et alii, 1969, 1970, etc.). They have enabled us to learn about pollen production in different countries; pollen production, like other apiary products, can be characterized according to its geographical origin, and controlled origin trademarks can be set up. In particular, these studies have provided relevant information about bee ethology, the bee-plant relationship, bee forage, the bee feeding preferences in any given phytosociological environment (Ricciardelli D'Albore and Persano Oddo, 1978). Furthermore, they have discovered the role that bees play in the pollination of some agricultural crops and of some ornamental and spontaneous species; they have also studied the bee's important role in pasture protection and in environmental protection in general.

These studies do not only deal with domestic bees, but also with other insects that can be of use to man.

To classify these insects as mono/oligo/polyleptic (this classification depends on whether the insects visit one, a few or many vegetable species of one or more families), it is necessary to analyse the pollen found in their hives; to this end it is sufficient to extract the food deposited close to the eggs by the queen bee. The undertaking of any programme in which insects are raised and then used for pollinating agricultural crops, requires a knowledge of the bio-ethology of those particular insects. It is possible to know where they gather pollen by collecting it from their hair and washing it with ethylic ether to prepare another specimen or by studying the different kinds of pollen contained in the ingluvia or in the whole intestine. Many pollinator insects (especially the solitary ones) do not have leg cavities in which to store pollen and so they collect it on their body or on their legs (for example Andrenidae) or on their abdomen (for example Megachile). These studies can also be applied to non Apoidea insects (Coccinellidae) to observe what kinds of food they gather (in certain situations ladybirds stop eating Rinchota and feed on pollen and fungus spores). Pollen analysis can also prove useful in environmental monitoring studies. In this case dead bees (eventually the beehive products) in a certain territory are analysed to see whether there are substances toxic to man as well as for gatherers (plant protection products, heavy metals, etc.) in that environment. The method basically consists in recording the worker bees' weekly mortality rate; when this rate exceeds a given limit, the dead bees are analysed to see if there are any toxic residues. Further information can be obtained by analysing the pollen stored in the hive or by collecting pollen from their hair and body and washing it with a solvent; in this way it is possible to trace the routes used by the gatherers and to identify the possibly polluted feeding areas. In conclusion, a melissopalynological study of the pollens gathered by insects can be very useful to both pure and applied research (Ricciardelli D'Albore, 1991).