Suppose that zonosemata flies whose own wings had been clipped and reattached were attacked more frequently than untreated zonosemata flies. How would this result have affected the reliability of the other experimental results?.