Melissopalynology - Pollen Analysis of Honey 

Melissopalynology, or the pollen analysis of honey, involves identifying and counting pollen grains in honey to determine the plants from which the nectar was collected. This method is primarily used to combat fraud and incorrect labeling of honey. For instance, monofloral honey, derived predominantly from a single plant species, is often more valuable than polyfloral honey, which comes from a mix of plants. Melissopalynology provides critical information to verify such claims and ensure product authenticity.

Among the various technologies for pollen analysis, DNA-based methods, such as PCR amplification of plant-specific markers and high-throughput DNA sequencing, provide superior sensitivity, specificity, and quantification. However, the standard method of morphologically identifying pollen species through light microscopy remains widely used. The following is an example of how I used this technique with my own honey.

I sampled a drop of honey from the surface of a jar harvested on August 6, 2023. At the time of sampling, the jar had been sitting for 484 days. Over time, pollen tends to float in honey, so this sample contained a significant concentration of pollen.*

*Pollen concentration is often very low in recently harvested honey, likely because it takes time for the grains to rise to the surface. In such cases, centrifuging a diluted honey sample would be helpful; however, I do not have access to a centrifuge.

The honey primarily originated from nectar collected by bees between mid-May and mid- July 2023. During this period, the most common plants visited by the bees, based on the number of pollen chunks dropped in the beehives, were:

• Blackberry: 31%

• Unidentified plant: 12.5% (resembles maple pollen, but I have not found a maple tree that blooms this late)

• Tulip poplar: 11.3%

• Chestnut: 8.8%

• Hawthorn: 5.9%

• Clover: 3.8%

• Sumac: 2.7%

Microscopic images of these seven species (taken from bee pollen chunks) are shown below. Note the variations in pollen grain shape and size within each species. For example, a small hawthorn pollen grain may resemble a large blackberry pollen grain, while a small blackberry pollen grain may resemble a large clover pollen grain. Additionally, pollen from poor-quality samples, such as wet pollen chunks, may appear rounder or more triangular rather than their original spindle shape, as seen in the blackberry pollen below.

The photos below show x800 magnification images of the honey sample. Each pollen grain is labeled as follows:

• C (Chestnut): Small, oblong pollen, approximately 0.02 mm long.

• T (Tulip Poplar): Relatively large, oval pollen, about 0.05 mm long, with a grainy texture.

• S (Sumac): Relatively large, oblong pollen, about 0.04 mm long, with a yellowish, grainy texture.

• ? (Other): All other pollen types, including round, oval, or oblong grains measuring 0.03–0.04 mm in length. These may include blackberry, an unidentified plant, hawthorn, or clover.

On a separate occasion, I took 50 random photos of the same honey sample and counted a total of 252 pollen grains. Of these, 175 (70%) were unidentified but likely included pollen from blackberry, hawthorn, and clover. The remaining grains were identified as 58 (23%) chestnut, 10 (3.9%) tulip poplar, and 9 (3.6%) sumac.

I was able to identify three species in this honey sample, based on my knowledge of its origin.

If I only knew this honey was from North America, I would still have identified 'T' as tulip poplar. However, 'S' could be citrus or related plant, and 'C' could represent any small oblong pollen, such as spiraea, hydrangea, and birdsfoot trefoil. Based on this, I could deduce the following:

• The honey originated from an area where tulip poplar grows, which excludes Zones 1–3, arid regions, high-altitude areas, and very tropical regions.

• The honey was harvested after the tulip poplar flow in early summer.

• The absence of easy-to-identify pollen species, such as sourwood, fireweed, pumpkin, linden, jewelweed, sunflower (or aster or goldenrod), and dandelion, might help further eliminate certain areas or seasons.

If no information on honey's origin was available, 'T' might represent a plant other than tulip poplar with similar pollen characteristics. In this case, the only conclusion I could draw is that the honey is not from areas or seasons where sourwood, fireweed, pumpkin, linden, jewelweed, sunflower (or aster or goldenrod), or dandelion are expected to be the major source plants.

1. Identification of Monofloral Honey

If the species of interest has unique pollen shapes or sizes, monofloral honey can be identified. Examples include sourwood, fireweed, tulip poplar, and linden.

2. Identification of Monofloral or Polyfloral Honey in Agricultural Areas

When honeybee hives are located within large, uniform farmlands without nearby related wildflowers, the floral source may be identifiable. Examples include canola (and other Brassica species), sunflower, and pumpkin.

3. Identification of Polyfloral Honey in Specific Places and Bloom Periods

If you know precisely which plants were in bloom during the nectar collection period, polyfloral honey may be identifiable, though certain limitations apply (as discussed in the example above).

1. Detecting Adulteration with Sugar Water or Corn Syrup

Honey containing very few pollen grains does not necessarily indicate it is fake. This could occur if the honey originated from honeydew rather than floral nectar or if pollen grains were lost during processing, such as ultrafiltration or pasteurization. Confirming adulteration requires chemical analysis of the sugar profile.

2. Determining the Country of Origin

While the dominant plant species contributing to honey vary by region, overlaps and similar-looking species can make identification challenging. For example, it may be relatively easy to distinguish honey from a Canadian farm from one produced in Vietnam (which would contain tropical species). However, verifying whether a "Made in USA" honey contains 50% honey from China would be difficult using pollen analysis alone.