Feather pecking is a problem on both economic and welfare grounds. There are several possible ways that feather pecking may spread. The simplest way is that one hen may damage the feathers of another hen, and another hen may find the damaged feathers an attractive pecking target. The objective of this experiment was to determine if damaged feathers were feather-pecked more than undamaged feathers on the same body area, and to determine whether some types of feather-body area manipulations were preferred over others as pecking stimuli. Manipulations of this experiment involved damaging the feathers on the rump, tail or belly of different hens with two or three levels of severity of manipulation. Sixteen groups of hens between 26 and 28 weeks were observed. Damaged feathers received significantly more severe feather pecks than undamaged feathers. The feather-body area manipulations that received the greatest number of severe feather pecks were the tail feathers when they were cut very short, the rump feathers when they were trimmed, and the rump when feathers were removed. The results of this experiment suggest that feather pecking does indeed spread through flocks by damaged feathers becoming an attractive target for feather-pecking behaviour. This finding is important in that it suggests that feather pecking can potentially spread through a flock if there is at least one hen in the flock that will start to damage feathers.