09 November, 2017

Bee studies - what can we measure and what should we measure?

By Matt Allan

Find me on:

Ecotoxicology is a relatively new study, less than 50 years old. Broadly it is defined as the study of harmful effects of chemicals on ecosystems, including individuals, populations and the environment. In discussing the effects on bees, it is important to also embrace the wider aspect of the effect on the colony of these social insects. 

Bees_pollinating_a_flower.jpgDeveloping bee studies

Initially, ecotox studies on bees were carried out almost exclusively on honeybees, because a lot is known about them, and for thousands of years they have become accustomed to being handled, managed and observed. Our first assumption, which was not unreasonable at the time, was that the honeybee was a reasonable surrogate species; meaning that effects observed on honeybees were probably going to be similar on other bee species, genera and families.

That was a good starting point, but it is clear now that the different biologies and behaviors of the 20,000 plus bee species in the world mean that different approaches are necessary for different types of bees. Now the ecotox community is engaged in developing study methods for bumble bees and solitary bees. Recently, another exotic group has been added to the agenda, the stingless bees of Central and South America. The task is therefore, to agree on one or a few species from each of these groups which are considered as representative of the group as a whole, and refine test methods accordingly.

So far, the consensus in Europe is that ecotoxicologists use the European honeybee Apis mellifera, the bumble bee Bombus terrestris (with its two geographical subspecies), and the mason bee Osmia bicornis (or alternatively Osmia cornuta if on mainland Europe). For more information about bee studies, you can read the development in Life Sciences article ‘Innovations in ecotoxicology: it really is the bee’s knees’

Gross effects on individuals and colony

Initially, bee studies focused on what might be called the gross effects, such as mortality of individuals; mortality of the colony in social species; reproductive success; and offspring success. The pattern of studies is always the same. Firstly, the chemical in question is evaluated to determine whether there is any risk involved in its use. If so, it is tested in the lab on individuals and groups of individuals.

If the risk is shown to be negligible no further study is required, but if there is a question over the risk, semi-field (i.e. in cages) studies are carried out. If the results indicate that there is still significant risk, then field studies are necessary. Full field studies are notoriously complex and expensive – and often contentious – so are avoided if possible for regulatory purposes.

Calculating mortality rates

Looking at mortality first, the honeybee colony is a dream subject, because it works like a metronome to a rigid timetable, with very well defined behavior patterns, and great regularity in size among individuals. As an organism, it is easy (at least for the practised beekeeper) to read and extract meaningful data.

Mortality - what do we measure and how?

At the moment, we want to consider only adults, as impact on juvenile forms requires in effect a separate and more complex investigation. The loss of the queen can be deduced by cessation of egg-laying, if not the discovery of her corpse. However, loss of the queen is not frequent as her exposure to the chemical under investigation is limited.

Different biologies and behaviors of the 20,000 plus bee species
mean that different approaches are necessary for different types of bees


The death of adult workers, on the other hand, is the most obvious evidence of impact – the commercial or hobby beekeeper has no problem detecting dead bees at the entrance to the hive caused by poisoning. Gross mortality is easy to see, and corpses to count. However, our studies have to be capable of detecting smaller but still significant effects. Let’s consider this in a bit more detail.  

Detecting smaller but still significant effects

Adult workers may die while still carrying out hive duties, in which case they are exposed to product brought into the hive. Corpses are not tolerated inside the colony; they are carried out by housekeeping bees and can be caught in dead bee traps at the entrance, giving us a good count of deaths in the hive.

Deaths of foragers are a different matter, as we do not easily know how many bees are flying for nectar, pollen, propolis and water, and cannot judge how many were lost away from the hive. However, technology exists in the form of bee counters that is capable of counting honeybees in and out of a colony.

Therefore two separate valid endpoints in terms of mortality can be evaluated for honeybees; these are deaths of foragers and deaths of hive bees. Potentially, queen mortality makes a third endpoint.  Questions about drones cannot be ignored – as sexual individuals, their well-being is essential for population stability – a topic that will be revisited in a later blog post.

Measuring bumble bees

The life cycle of bumble bees differs from honeybees in that they form social communities for only part of the year and it this stage that is studied. Again, mortality impacts on the queen, nest workers, foraging workers and drones. For the time being, this blog post will only consider adults in this discussion.

The inside of a bumble bee nest can be a messy place. A dead bee, whether adult or juvenile, is not tolerated in a healthy colony. Initially, they are dragged to the perimeter of the nest and in due course carried out and discarded. However, there is considerable size variation in individual adults, so the type of dead bee trap used for honeybees is not particularly effective.

As routine, only dead bees are counted in the nest at three or four day intervals but have no knowledge of how many have been removed in the intervening period. When evaluating differences between controls and a toxic reference, there may well be a significant impact. Consequently, when looking for subtle impacts, this is not an adequate assessment method and forager mortality may be a more useful assessment tool.

To date, there has been little effort to gather precise data. In cages, it is recommended that cloth is spread over part of the enclosure (maybe 10% of the area) and the number of dead adults found on the cloth counted. Apart from the imprecision of the method, a dead bumble bee is a nutritious treat for ants, beetles and rodents, and a bee corpse does not remain long on the ground.

Improved approach to counting bee mortality

A better approach that should be adopted is using bee counters. By counting the numbers of bees that leave the nest and the number of bees that return to the nest, a direct count of mortality outside the nest can be obtained which is reliable and accurate. A further advantage of this method is that it is equally useful in the field situation, where bumble bee colonies are studied in the open. As before drones or juveniles but will be discussed in a subsequent blog post.

Considering other bee species

Solitary bees or mason bees are very easy to study for a host of reasons. When considering adult mortality, again focusing on females (and leaving the awkward males till later), it is easy to assess as every female roosts in a nest at night. Therefore, it is easy to count how many adult females are present every 24 hours.

Mortality is assessed as the reduction in the number of roosting females per day. Not only does this easily indicate gross impact, but it is also capable of detecting subtle impacts.

Future blog posts 

The discussion in this blog covered the impact on adults in the three different bee groups, subsequent blogs will address studies on juvenile forms and also consider sub-lethal effects.

For more information about bee studies, you can read the development in Life Sciences article ‘Innovations in ecotoxicology: it really is the bee’s knees’

Regulations and registrations: your guide to successful ecotoxicology studies on bees

About the author

Matt Allan, Atlantic Pollination

Matt is a pollination specialist who has been involved professionally with bees for over 30 years, working in the fields of education, advisory services, manufacture, rearing, and pollination, across Europe and the USA. Matt shares insights from the ecotoxicology arena and, specifically, on bee studies, where he reveals some highlights from the projects he is collaborating on with Envigo. If you would like more information about Atlantic Pollination's services please contact Matthew Allan. His other blog posts include;


Category // Crop protection, ecotox innovations, ecotoxicology, Matt Allan, "pollination", bees