Fig. 2. Division of labor and activity rhythms in free flying B. terrestris colonies. (A) Representative actogram for a forager. The numbers on the y-axis depict the day of observation. The height of the bars within each day corresponds to the number of scans in which the bee was active (see Materials and methods for details). Data for each day are double plotted to facilitate visual detection of rhythmicity. This individual has significant diurnal rhythms in activity. The bar at the bottom of the plot depicts the illumination regime. Open bar=day (sunrise to sunset); filled bar=night (sunset to sunrise). (B) Representative actogram for a nurse. Details of plot as in A. Hatched part of the illumination regime bar depicts the time of day for a bee that stays inside the constantly dark hive. This individual has no diurnal rhythms in activity. (C) Representative summary of the percentage of rhythmic nurses and foragers (Colony 2, session B). A higher proportion of foragers have significant circadian rhythms (Fisher's Exact Test, P<0.001; similar results were obtained for two additional colonies in session B and for colony 2 in session A). (D) Development of diurnal rhythms in young foragers and intermediates in free-flying colonies (pooled data from all three colonies). The proportion of bees with circadian rhythms was higher in session B (Fisher Exact Test, N=13, P<0.05). There was no similar development of rhythm in nurses. (E) Representative summary for the relationship between body size and the power of circadian rhythms (Colony 2, session A, R²=0.34, N=15, P<0.05). Similar results were obtained for an additional colony. Note that in this experiment foragers experienced day–night oscillations whereas nurses typically stayed inside the relatively constant environment of the hive.