The diagrams in the paper by Y. Namikawa and Y. Yamazaki with the above title (this Journal, 22, 169) are based on a fallacious argument. Since the distance run between bearings is one of the parameters a navigator must choose, it is useless to express the error as a multiple of it. Consequently their final recommendations are also invalid.

Instead, the error should be expressed as a multiple of the shortest distance off, a distance which is not affected by the pattern of observation chosen. The navigator must then choose the angles off the bow, or more correctly off the course made good, to obtain the best fix.