Fig. 3. Summer and autumn orientation of juvenile loggerhead turtles. Each dot within a circular diagram represents the mean angle of orientation for a single turtle during its 10-min trial. Triangles on the outsides of the circles correspond to the most direct routes back to the capture areas for each group of turtles. The broken lines represent the 95% confidence interval for the mean heading. (A) Orientation of loggerheads captured at locations east–northeast of the test site and tested between May and September 1998–2001. Turtles were significantly oriented with a mean angle of 83° (N=122, r=0.29, P<0.0005 V-test, 95% confidence interval ±25°). The confidence interval overlaps the direction to the capture area (74°). (B) Orientation of loggerheads captured at locations west–southwest of the test site and tested between May and September 1999–2002. Turtles were significantly oriented with a mean angle of 271° (N=11, r=0.43, P<0.05 V-test, 95% confidence interval ±76°). The confidence interval overlaps the direction to the capture site (255°). (C) Autumn orientation of loggerheads tested in October and November 1998–2000. Turtles were significantly oriented with a mean angle of 190° (N=15, r=0.61, P<0.005 Rayleigh test, 95% confidence interval ±33°). This direction coincides with the direction of orientation exhibited by wild turtles during their autumn migration.