The present research examined whether eyewitness identification lineups produced by law enforcement personnel are biased or suggestive. Experienced police officers were asked to construct two six-face photographic lineups, first using their usual (traditional) method, and second using an alternative method. The primary basis of the traditional method is that foils are selected based on their similarity to the target. The alternative method includes foils that are not only similar to the target but also similar to other foil faces in the lineup. Both types of lineups were shown to subjects who had not seen the faces before (mock witnesses) and were asked to guess the respective targets. The results showed that mock witnesses selected the targets significantly more often than expected by chance (1/6 probability) when embedded in the traditional lineups, thus demonstrating that these lineups were suggestive. Mock witnesses did not select alternative-method targets more often than expected by chance. These results indicate that foil selection procedures incorporating foil-to-foil similarily produce fairer lineups than those exclusively based on target similarity. Implications for forensic lineup construction procedures and for future research are discussed.