In the second of a series of experiments reported earlier (Day, 1973), it was found that the apparent misalignment of two oblique, collinear lines forming the arms of two 45 0 acute angles was no greater than that of the two oblique lines alone. It was therefore concluded that the acute angles do not themselves contribute to the misalignment illusion. A more recent experiment by Wenderoth, Beh, and White (1978) did not support this conclusion. It was found that apparent misalignment between a dot and an oblique line forming the arm of a 45 0 acute angle was greater than that between the dot and the oblique line alone. While these outcomes appear inconsistent, the two experiments were different in a number of ways. In Day's experiment, the display was horizontal, the method of adjustment was used, and the angles and lines were derived from the parallel. line figure shown in Figure lA. In the experiment described by Wenderoth et at., the display was vertical (the face of an oscilloscope), the random interleaved staircase method was used, and the angle, line, and dots were derived from the basic figure shown in Figure lB. However, since, except for the relative magnitude of misalignment for the acute-angle and single-line figures, the other results reported by Wenderoth et at. were in good agreement with Day's, it seems unlikely that the different displays or procedures were implicated. It is more likely that the differences between the members of the two pairs of figures were involved. The figures used by Day (1973) are shown in Figures 2A and 2B and those used by Wenderoth et at. (1978) in Figures 2C and 2D. It can be seen that, in Day's experiment, one figure consisted of two acute angles, each with a vertical arm slightly shorter than the oblique and the other of two oblique lines. In the experiment by Wenderoth et aI., one figure consisted of a single acute angle and a dot with the vertical arm of the former about three times longer than the oblique. The other figure consisted simply of an oblique line and a dot. The purpose ofthe experiment reported here was to establish whether the inconsistency between the two earlier experiments was attributable to differences between the two pairs of figures. Therefore, the two pairs, Figures 2A and 2B and Figures 2C and 2D, were essentially the same as those used originally.