Route recognition in the homing pigeon, Columba livia JESSICA MEADE, DORA BIRO & TIM GUILFORD Animal Behaviour Research Group, Department of Zoology, University of Oxford (Received 22 October 2004; initial acceptance 18 November 2004; final acceptance 28 November 2005; published online 7 September 2006; MS. number: 8310R) While homing pigeons are known to use familiar visual landmarks to recognize release sites, it is less clear whether they continue to attend to visual cues along the homeward flight. To address this question, we used precision GPS technology to track pigeons along their homeward route and later released them from sites that they were known to have flown over. The group of birds that were aerially familiar with the chosen sites homed significantly more efficiently than did their yoked (na € ıve) counterparts. We found no evidence that birds were able to recognize sites that they had previously flown over from the substan- tially different viewpoint perceived at ground level. The results imply that the birds were able to recognize some aspect of the chosen sites after release and used this information to home more efficiently, although the nature of these cues is not clear. Ó 2006 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. The influence of visual landmarks on pigeon navigation has been a controversial issue throughout many years of research. Initially, visual landmarks were thought to be important for homing, as pigeons are highly visual animals (Hitchcock 1950; Griffin 1955), although other authors were sceptical about this hypothesis (e.g. Mich- ener & Walcott 1967). Schmidt-Koenig & Schlichte (1972) showed that some pigeons could return to within a few hundred metres of their home lofts while wearing frosted contact lenses, which allowed a view of the sun but no detailed visual features. This result led researchers largely to ignore the use of visual landmarks in pigeon homing for many years. Researchers have since accepted that pigeons have much redundancy in their homing sys- tems, and that vision is important, particularly when pigeons are homing in the familiar area (Wallraff 1991). Although much evidence now demonstrates the impor- tance of familiar visual landmarks to aid homing over short distances (Braithwaite & Guilford 1991; Braithwaite & Newman 1994; Burt et al. 1997; Biro et al. 2002, 2003), these experiments only investigated familiar visual land- mark use at release sites. This limitation was due to a lack of technology, in that release sites were the only areas that experimenters could be sure that a pigeon had actually visited. Miniature GPS loggers (von Hu ¨ nerbein et al. 2000; Steiner et al. 2000) have now made it possible to record a pigeon’s entire homeward path. This provides experimenters with a powerful tool to investigate the cues attended to by pigeons along the homeward route. Recent experiments releasing pigeons repeatedly within a familiar area (Biro et al. 2004; Meade et al. 2005) have shown that pigeons develop highly stereotyped routes, which implies that they rely upon familiar visual landmarks throughout the homeward journey and not just at the release site. We extended an experiment performed by Braithwaite (1993), which involved releasing birds from sites that they had previously flown over en route home. We used GPS technology that allowed precise determination of ae- rially familiar locations for individual birds to compare the effect of aerial familiarity on homing efficiency with per- formance by yoked (na € ıve) controls. We also adopted the standard preview paradigm (e.g. Braithwaite & Guilford 1991) to determine whether aerial familiarity can trigger visual recognition from the ground. METHODS Subjects Subjects were 16 homing pigeons reared at the Univer- sity Field Station, Wytham, Oxford, U.K. All were at least 2 years old, weighed at least 450 g, were successful homers and had been used in previous experiments. Birds were kept in a social group of about 140 individuals between two lofts (ca. 6  2 m and 2 m high). Plentiful perches and nestboxes were available and all birds had access to an outside aviary. Food (a commercially available multi- grain mixture), grit and water were provided ad libitum throughout the study. All subjects were fitted with a Velcro strip (30  70 mm), attached to their back with flexible Correspondence: T. Guilford, Department of Zoology, University of Oxford, South Parks Road, Oxford OX1 3PS, U.K. (email: tim.guilford@zoo.ox. ac.uk). 975 0003e 3472/06/$30.00/0 Ó 2006 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. ANIMAL BEHAVIOUR, 2006, 72, 975e980 doi:10.1016/j.anbehav.2005.11.029