Reproductive conflicts in honey bees: how new queens are chosen, and worker egg laying is suppressed

Worker bees are usually sterile, and reproduce only via their queen. But evidence is growing that workers influence their reproductive fitness in subtle ways. When a queen dies, some genotypes are reared as new queens more often than others. Whether this is due to nepotism, or to royalty genes which make some bees more attractive than others for rearing as queens is being assessed. Occasionally too, some workers escape sterility and lay fully viable male eggs. How this cheating is kept under control, and how this control sometimes breaks down is being determined using microsatellite analysis.

Progress

We have shown that in queenless colonies of honey bees, the selection of worker larvae which are developed as queens is significantly non-random with respect to subfamily. (Results from one such colony are shown below, with subfamily D being particularly successful in becoming queens). This is clear evidence of a genetic component to the relative reproductive success of workers. The phenomenon indicates either nepotistic behaviour or that the larvae of some subfamilies are more attractive for development as queens.

A colony in which worker bees were laying eggs (which we call 'anarchistic' behaviour) was purchased from a beekeeper in March 1995 (beekeepers can identify anarchistic colonies by the presence of drone brood located in honey supers to which the queen cannot gain access). In October 1995, this colony's workers again laid eggs, and about 1000 worker-laid males were produced during November. (This provides evidence that the phenomenon has a genetic component since it is repeatable, but that environmental effects are also important.) Queens were raised from this colony and instrumentally inseminated. We now have a line that shows anarchistic behaviour with high frequency available for experiments. A naturally occurring anarchistic colony was studied in detail in 1996 using microsatellites.

In this colony, the great majority, if not all, males produced were offspring of workers, not the queen. Over 85% of the egg-laying workers were of a single subfamily. Upon removal of the queen of this colony, workers commenced egg-laying in worker-cells within days, whereas workers in 3 control colonies similarly manipulated did not do so. Interestingly, the first bees to commence egg laying were of a single subfamily, but not the same one as had been previously identified as anarchistic. Later, bees of other subfamilies commenced egg-laying. This shows that the anarchistic behaviour is not an epiphenomenon of queenlessness. We initially thought that anarchistic behaviour might have resulted from selection for rapid ovary development by workers following queen loss. Our 1996 observations show that the situation is more complex.

Observation hives studies have shown that anarchist workers do not form ‘courts’ as do queens or queenless workers, nor are they attacked by non-laying workers. Therefore their mandibular pheromonal bouquet does not appear to be queen-like. Anarchistic workers undergo a normal but highly precocious behavioural ontogeny.

We are very excited about this project, and is well funded for the next three years.