Bumble bees often restrict their foraging behaviour to the flowers of two p
lant species on a single foraging trip ("a major and a minor" sensu Heinric
h), perhaps because learning additional flower-handling techniques interfer
es with their ability to recall previously learned flower-handling skills.
This hypothesis was tested with bumble bees (Bombus impatiens) foraging on
arrays of artificial flower types. Test bees were first trained on one flow
er type, then retested on the same flower type after learning one, two, or
three different types. The number of flower types selected by bees on a mix
ed array consisting of several types was also tested. Interference effects
increased with the number and complexity of interfering flower types: a 0%
increase for one simple flower type; 278% for two simple flower types; 357%
for three simple flower types; and 565% for two complex flower types; swit
ching among more than two flower types caused substantial interference. On
the mixed array of three flower types, none of the 20 foragers tested rando
mly visited all three types, 67% primarily visited two types, and 33% prima
rily visited one type. These findings suggest that the bees limited the num
ber of flower types to two because of the substantial costs incurred when a
third flower type was included in their foraging repertoire.