World wide the International Standards for Honey are laid out in the Codex Alimentarius (administered by the "Joint FAO/WHO Food Standards Programme") to which New Zealand is an active signatory. The Codex standard for honey can be downloaded in PDF form here. The relevant section for monofloral honeys in the Codex Alimentarius is:
Section 6 Labeling
Section 6.1 The Name of the Food
"Honey may be designated according to floral or plant source if it comes wholly or mainly from that particular source and has the organoleptic, physicochemical and microscopic properties (our emphasis) corresponding with that origin."
Where honey has been designated according to floral or plant source (6.1.6) then the common name or the botanical name of the floral source shall be in close proximity to the word "honey".
It is clear that some measurements have to be applied to a specific honey to determine if it is in fact what is claimed to be on the label. To this end we routinely measure the following:
All monofloral honey types we deal with have a limited range of values for each of these measurements. By taking each measurement and comparing it with the average value and standard deviation of the target honey type from our database of over 26,000 samples, we are able to determine, with excellent repeatability, if a honey type is true to label.
Colour of Honey - Pfund Grader
Honey colour is measured on a "Pfund Grader". The scale for this is called the "Pfund" Scale and is measured in millimetres. The scale is actually a metric ruler measuring the point along a calibrated amber glass wedge where the sample (placed in a glass wedge shaped trough) matches the amber wedge. The scale starts at 0 mm (colourless) and finishes at 140 mm (black). Some common terms describing the colour of honey actually have specific ranges on the Pfund Scale. These also vary slightly from country to country with the USA, Canada and Australia all having slight variations. The US scale is :
In many cases we have over 1,000 samples on file for a particular honey type and are able to compare the average colour plus Standard Deviation (SD) of the colour with that of a sample under analysis.
There are several papers in the scientific domain that refer to microscopic analysis and pollen analysis in particular that give broad indications of required levels of pollen for a single floral source statement to be made. In New Zealand, a paper titled "Pollen Analysis of New Zealand Honey" was published in the New Zealand Journal of Agricultural Research in 1985. This paper gives required levels of pollen for specific New Zealand honeys. It is therefore the source document that would be used in applying section 6.1.4 from the Codex standards.
Pollen Analysis is based on identification of pollen in a sample of honey but may also extend to identification of other particles in the honey e.g. in the case of honeydew.
Pollen finds its way into honey by two routes.
Firstly, and most importantly for pollen analysis, pollen naturally falls into the nectar at the flower. This is collected by the bees and taken back to the hive to be ripened into honey. Different plants produce a wide range of pollen sizes and types. This along with the structure of many flowers affects how much pollen on average falls into the nectar and (thence into the honey) for a given plant species.
A good example of how flower structure can affect the incidence of pollen in nectar is New Zealand Fuchsia (Fuchsia excorticata). Here the anthers and pollen hang down and are physically separated from the nectaries.
Fuchsia is primarily a bird pollinated plant and the flower structure is adapted to putting pollen onto a bird collecting nectar. Pollen falling off the anthers falls away from the nectaries and bees collecting nectar do not come into contact with Fuchsia pollen. It is an uncommon pollen in New Zealand honey, in spite of Fuchsia being considered a good Spring nectar source.
In the honey World, three classifications have been proposed. Honeys are deemed to be Under, Normally or Overrepresented. Under represented is deemed to be nectar sources that produce honeys with less than 20,000 pollen grains per 10 grams of honey. Normally represented are those with 20,000 - 100,000 pollen grains per 10 grams of honey and over represented honeys have more than 100,000 pollen grains per 10 grams. When looking at the percentages of pollen in a honey sample, the classification the various nectar sources fall into has to be taken into account. For a honey to be identified as a monofloral honey type, e.g. "clover", "manuka" etc. the following levels should typically be met:
Under represented pollens, a level of greater than 20% of the target pollen must be identified,
Normally represented pollens, a level of greater than 45 % of the target pollen,
Over represented pollens, a level of greater than 70% of the target pollen is require
The other way that pollen can arrive in honey is from contaminating sources. This typically happens during the extraction process when pollen in the frames (collected and stored by the bees for a protein source) is removed with the honey.
This pollen may not be the same source as the nectar source and is collected independently by bees foraging specifically for pollen as opposed to nectar. In a routine pollen analysis the total pollen content of the honey should be measured to see if there is a "masking" effect from contaminating pollen at time of extraction. The total pollen figure should not be a magnitude higher than normal for the target species.
In fact the simplification of "Under, Normally and Over" represented honey types does not cover all possible situations. Many plants do not end up with significant pollen from the source plant in the nectar. The Fuchsia example given above is an extreme example with little if any pollen getting into the nectar. Nodding Thistle is another example where we believe the structure of the flower, (long thin florets tightly packed together) and the large size of the pollen grains (around 60 microns) prevent much NT pollen from getting into the nectar. We have found that most NT honey only has around 500 - 1,000 pollen grains per 10 grams. This is an extremely low figure and means that as a percentage of other pollen types that get into the nectar from other sources (or from other nectar sources collected with the [predominantly] NT nectar) NT is very low. This is often less than 10%.
Pollen analysis is a valuable tool for honey identification, however it should be pointed out that this is a complex subject and is always used in conjunction with other information before a final determination is made.
Additional References on Pollen Analysis.
Measuring conductivity is an indirect way of measuring the mineral content of a honey. As sugars in solution (the main component of honey) are poor conductors, any minerals present in the honey aid the ability of honey to conduct electricity. For honey, this is expressed as ohms/cm x 10-4 taken in a solution of honey and distilled water (at 20°C) where the dry matter of the honey makes up 20% of the weight of the solution.
Typically flower honeys are less than 1.5 while honeydew honeys are greater than 8.5. There are of course some exceptions to this and this feature is an aid in the identification of these honey types. Also blends of honeydew honeys and flower honeys will give an intermediate value. Manuka is one flower honey that has a much higher than usual conductivity for a flower honey.
As the largest component of honey, sugars are an important part of honey identification. The two main sugars are fructose and glucose and the ratio of these two sugars can be a useful determinant for identification. We use HPLC (High Performance Liquid Chromatography) as our analysis method, and while relatively new to our laboratory, is giving us new found insights into the honeys we work with.
While fructose and glucose are the major sugars, there is a vast array of more complex sugars found in honey and each honey has its own spectrum of sugars that are typical to that honey. Of interest in honeydew
is the presence of oligosaccharides