Fig. 3. Evaluation of homing directions in experiment I. For each trial, the homing direction was measured in two coordinate systems. (A) The animal's position at the end of the homing trip, when it reached the 15 cm-wide peripheral annular zone of the arena (dashed circle) was coded with respect to the real arena floor, which was subdivided into 10° sectors. The 0° reference direction was the radius vector pointing from the centre of the arena to the standard nest location; positive values corresponded to clockwise angles. As the return to the nest (curved path) does not start at the arena centre, this measure entails a bias (see text). (B) The animal's orientation was therefore assessed by a second measure, at a distance of 35 cm from the start of the return, with a mobile grid (dotted circle), also subdivided into 10° sectors. The grid was centred on the start of the return at P2 (or P1), and its 0° reference direction was aligned with the standard position of the nest entrance.