2.1.1. When to sample?
A honey bee colony is a complex super organism with changing features in response to (local) seasonal changes in the environment. Average age increases, for example, in colonies in the autumn in temperate regions, because of the transition to winter bees. Age-related tasks are highly plastic (Huang and Robinson, 1996), but after a major change of a colony’s organisation it can take some time before the division of labour is restored (Johnson, 2005). Immediately after a colony has produced a swarm, for example, bees remaining in the nest will have a large proportion of individuals younger than 21 days, lowering the average age of bees in the colonies. Over time, these bees will become older and the average age of bees in the colony will increase again. Therefore, it is recommended that if the aim is to have an average / normal / representative sample with respect to age structure, one should only sample established colonies that have not recently swarmed. The same is true for recently caught swarms, because brood will not have had enough time to develop, and one could expect rather an over-aged structure. Age polyethism in honey bees and its implications for the physiology, behaviour, and pheromones is discussed in detail elsewhere (Lindauer, 1952; Ribbands, 1952; Lindauer, 1953; Jassim et al., 2000; Crewe et al., 2004; Moritz et al., 2004).
Furthermore, physiological variables in individual bees (and in pooled samples from a colony) can change over time when these parameters are, for example, related to age of bees or presence of brood. Moreover, build-up of vitellogenin takes place in the first 8-10 days of a bee’s adult life and then decreases, but is much faster in summer than in winter when no brood is present and bees are on average older (Amdam and Omholt, 2002), affecting averages of individual bees of the same age, but also averages of pooled samples. Nosema apis, Paenibacillus larvae, and Melissococcus plutonius are examples of organisms with bee age-related prevalence in colonies. N. apis infections are not microscopically detectable in young bees; after oral infection it takes three to five days before spores are released from infected cells (Kellner, 1981). P. larvae and M. plutonius can be detected in and on young bees that clean cells (Bailey and Ball, 1991; Fries et al., 2006; Carreck et. al., 2013). Depending on the disease, higher prevalences can be found in colonies with relatively old and young bees, respectively. Furthermore, seasonal variation in pathogen and parasite loads may also affect when to sample; for example, screening for brood diseases during winter might be inappropriate.