Extensive and rapid food transfer occurred between colonies of honeybees when 2 colonies were confined in each of 3 outdoor flight cages. Using visually distinct genotypes (cordovan and Caucasian) we could distinguishbetween worker bees of the 2 colonies within each cage. One colony in each cage was fed with syrup containing rhodamine dye. In samples of bees taken from clusters outside the hives after 24 h of treatment, dye was found in the digestive tract of 27 - 56% of workers from the untreated colonies; after 48 h it was present in 53 - 74%. Of all worker bees samples from inside the untreated colonies, 45% contained dye after 68h. Cells in all combs from untreated colonies had a deposit of dye, and food surrounding the larvae was alos dyed. After 5 days only 0.5-3.6% of the worker bees in untreated colonies had drifted thre from treated colonies; 11 out of 1200 bees were observed enteringthe wrong hive. Although 93% of the drifted bees had dye in the digestive tract, driftwas much less frequent than food sharing between clustered bees. Therefore, the primary source of transfer of dye between colonies was food sharing by bees clustered in the flight cages and not from bees drifting between colonies. Based on these data, we recommended that, in studies using honeybees in outdoor flight cages, only 1 colony be placed in each cage.