The Boström-Kugelberg pseudo-isochromatic plates. How efficient is the third edition?

The third edition of the Boström-Kugelberg pseudo-isochromatic plates was printed to the best possible visual match with the second edition. Slight colour differences between the 2 editions initiated this study where 72 colour defectives and 57 normal trichromats were tested with the plates and classified with the aid of Nagel's anomaloscope. Of the defectives, the plate test disclosed all but two anomalous trichromats. One normal subject was falsely classified as defective. The plates had to be shown twice according to the instructions since otherwise 13 of the 53 normals were misclassified. Three plates were shown to have low sensitivity; one is best used as a demonstration plate, and the other two can be deleted without lost efficiency. Forty-eight of the colour defectives were also studied with Farnsworth's lantern. The correlation between the numbers of plate and signal errors was weak. All subjects who failed the lantern test made several plate errors, but 15 subjects who passed the lantern test, made two or more errors on the plate test. The outcome of the plate test does not give evidence of the subject's lantern test performance.