Aphis fabae (Homoptera: Aphididae) foundresses were shown to have a clustered distribution on leaves of their primary host, Euonymus europaeus. Two field experiments were carried out to determine the costs or benefits of this clustering behaviour to aphid clones. The first experiment, in spring 1996, excluded predators from colonies formed by single fundatrices or clusters of three. No density-dependent effects on colony growth were observed early in the season, but the total production of migrants per foundress was significantly lower for clustered foundresses. This demonstrated that there was no synergistic effect of group feeding on clonal performance. All colonies were initiated on randomly selected leaves and all grew successfully, therefore clustering was not due to a shortage of feeding sites. The second experiment, in spring 1997, tested whether natural enemies cause selection for selfish herding by the foundress generation. Colonies were initiated by one or three foundresses, but half were left unbagged. If natural enemy attack is a selective force inducing clustering behaviour in foundresses, the costs of clustering should have been reduced or negated in unbagged colonies. Colonies grown from clusters produced significantly fewer migrants per foundress, irrespective of bagging treatment. Other potential factors causing clustering behaviour are discussed.

Primary host Aggregation Clonal fitness Aphis fabae Natural enemies

Aphididae population distribution Behavior insect Aphis fabae colony growth behavioral ecology Homoptera migration Euonymus europaeus experiment