Some published figures relating to the incidence of swarming have been examined, and records obtained in the routine examinations of colonies in honeyproducing apiaries have been analysed. It is concluded that in these apiaries between 10 and 40% of colonies would swarm in an average year, if they were given excess hive space and otherwise left alone. The proportion of colonies which produced queen cells varied from year to year and from one apiary site to another. This variation was presumably due to environmental factors, but it was substantially independent of those factors which determined honey yield. There was no evidence that it was due to the effects of conditions in the years when the queens were reared. After the end of June, at least, the tendency for colonies to produce queen cells was markedly less with queens of the current year than with queens of the previous year, and was probably also less with queens of the previous year than with still older queens. Queen cells were most frequently observed in colonies in the latter half of May and in June and July. The mean time of queen-cell production varied from year to year. Many colonies which began to rear queens eventually ceased to do so with no other treatment than removal of queen cells. Queen rearing did not begin and end in all colonies at the same time; many colonies began rearing queens after others had stopped.