2i6 SECTIONAL ADDRESSES 



learned to accept an equation as correct which looked wrong to him. 

 One of the investigators, himself a deuteranomalous trichromate, to put 

 this possibility to the test, practised daily, and learned quite soon to 

 recognise the normal equation, although it looked quite wrong to him. 



These examples suffice. to show some of the difficulties encountered in 

 the detection of colour anomaly. Tests must be so devised that the 

 person tested cannot avail himself of any secondary aids, but has to rely 

 solely on his own colour sensations. 



The spectrometer is undoubtedly the most fundamental test of colour 

 vision, but it is seldom available for practical purposes, and the majority 

 of tests, apart from those in scientific laboratories, are carried out either 

 with some kind of lantern test or some kind of pigment test. There are 

 various types of lantern tests available. The Edridge- Green Colour 

 Perception lantern is the only one which has been used by the writer, 

 and its advantages have been referred to elsewhere.^'' A newer model 

 is the Board of Trade Lantern Test, recommended with modifications 

 by the committee set up to consider Colour Vision Requirements in the 

 Royal Navy. In using this the eyes have to be dark-adapted for 

 15 minutes. The Giles-Archer Perception Unit is also a new and simple 

 model, and it, too, requires the eyes to be dark-adapted. These lantern 

 tests enumerated, and many others, have the advantage that coloured lights 

 are used instead of pigments, which brings conditions of testing nearer to 

 everyday conditions in the services, the railroad, navigation and aviation. 



Certain pigment tests have also been extensively used, and it is these 

 I should like to discuss in some detail. In some of these tests, it is 

 puzzling to find mistakes made sometimes by individuals with normal 

 colour vision which should only be made by colour blinds. The printing 

 of the tests may be partially to blame, but it must be recalled that decisions 

 are constantly being made on the results from these tests, and therefore it 

 is essential to recognise which responses are diagnostic and which can be 

 ignored. In order to reach a valid basis for diagnosis, I have given a 

 battery of tests under constant conditions of distance and illumination to 

 an unselected group of about 340 candidates applying to be accepted as 

 apprentice printers, exclusive of colour blinds. Their responses, 

 therefore, to the tests may be compared with the responses of a group 

 of colour blinds tested under the same conditions. This normal group 

 acts as a control group against which the results for each test used can 

 be evaluated at its proper worth. 



Owing to exigencies of space and time it will not be possible, nor 

 desirable, in this address to discuss the details of the results obtained 

 with all the tests which have been used. I have, therefore, decided to 

 indicate the type of investigation under progress by giving the results 

 from two of the tests only. The two tests selected are the Ishihara Tests 

 for Colour Blindness and the Mosaic Plates devised by Schaaff. 



I. The Ishihara Tests for Colour Blindttess (5th Edition). 



The Ishihara test is composed of a number of pseudo-isochromatic 

 plates in which coloured numerals appear on coloured backgrounds. 



^' Collins : Colour Blindness. 



