In three experiments, 392 introductory psychology students witnessed staged crimes and attempted to identify criminals from photographic lineups; both the guilty and similar looking but innocent suspects were pictured. Lineup attire was manipulated in three ways, however the identification data revealed that the rate of identifications of the guilty party was not influenced significantly by lineup attire. The study revealed that biased lineups will not increase correct identifications, but will increase false identifications relative to other procedures. Eyewitnesses who selected the 'suspect' clothing were significantly more accurate in their identification of the person than eyewitnesses who failed to select the suspect clothing. The implications of the findings for police identification procedures are discussed. Table, 20 references. (Author abstract modified).