83 
1921-22.] New Method of investigating Colour Blindness. 
In spite of my earlier tests, however, this observer maintained that his 
colour vision was good, but admitted that under certain circumstances he 
was unable to recognise green. The diagram shows that his view is quite a 
reasonable one ; he gets 64 patches, which is not far from normal. He gets 
All the observers represented above make very serious errors in naming and matching colours, 
and would have been classified as totally red or totally green blind by the old Board of 
Trade test. No. 11 thinks Vandyke-brown a perfect match for vermilion and Prussian 
blue. 
rather a low number of steps, 6J, on the red-green side, and a relatively 
high number, 4, between peacock-blue and green. The point in the spectrum 
where the colour changes rapidly is in his case shifted from yellow into 
green. It is ridiculous to call an observer of this type ‘‘colourblind.” 
“Colour different” is a more suitable name. It is owing to the colours in 
common use not being strictly monochromatic that he makes mistakes; the 
greens in common use include wide ranges of spectrum which change slowl}'’ 
