734 REroKX — 1892. 



cyan-blue, yellow and ultramarine-blue, greenish-yellow and violet, tlie comple- 

 ment for pure green being found not in any liortiogeneous ligbt, but in purple — a 

 mixture of red and violet. The complementary colours may be arranged in a 

 circle, with the complementaries in each pair placed opposite one another. Of 

 course, the circle cannot be completed by the colours of the spectrum ; purple 

 must be added to fill in the gap between the red and violet. Helmholtz found no 

 constant ratios between the wave lengths of homogeneous complementaries ; and it 

 is a striking fact that, while a mixture of the green and red, or of the green and 

 violet, undulations gives rise to a sensation such as could be produced by rays of 

 intermediate wave length, no such etiect follows the mingling of rays from oppo- 

 site sides of the spectrum. Pure green, with a wave length of 527 millionths of 

 a millimetre, marks the division between the right and left sides. The mixture 

 of blue from the one and yellow from the other side does not produce the inter- 

 mediate green, but a sensation of white. A mixture of blue or violet and red 

 produces not gieen, but its complementary — purple. On the trichromic theory, 

 the sense of wliite produced by the mingling of any of these two colours is simply 

 regarded as the result of a balanced stimulation of the red, green, and violet terminals. 



But the anatomical form given by Helmholtz to Young's theory of primary 

 colour sensations is beset with serious difficulties. It implies the existence of 

 three sets of terminals in the retina, and these must all be found in the central 

 part of the yellow spot where cones alone are present. Three sets of cones there 

 would be necessary to respond to tlie ]'ed, green, and violet light, and a colourless 

 pencil of light could not be seen uncoloured, unless it fall on three cones, which 

 ■we know is not the case. Therefoi'e, if there are three diflerent terminals, they 

 must, in the human retina at all events, be found in every single cone in the 

 yellow spot. In a single cone there might be three sets of fibrils capable of simul- 

 taneous stimulation in ditt'erent degrees, but it seems impossible to suppose that the 

 different vibrations started in the same terminal could be kejjt discrete, and be 

 transmitted to the brain through tliree d liferent fibres in the optic nerve, even if 

 the nerve contained a suihcient number of fibres for the purpose, which we know is 

 not the case. 



The phenomena of colour-blindnpss also offer great diihculty. In several cases 

 of apoplectic seizure it has happened that the ceuti'e for vision on botli sides of 

 the brain has been completely or partially paralysed by the extravasated blood. 

 In such cases the sense of colour may be entirely lost either ibr a time or per- 

 manently, while the sense of light and form remain, altliough impaired. The 

 loss of colour sense in some cases lias been found complete in both eyes ; in most 

 of the recorded cases the loss of colour sense was limited to the right or left halves 

 of both eyes ; that is, if the lesion aflijcted the vision centre on the right side of the 

 brain, the right halves of both eyes were blind to all colours. That illustrates the 

 well-known fact that a sense of light does not necessarily imply a sense of colour. 

 The colour sense probably involves a more highly refined action of the sensory cell 

 than the mere sense of light and form, and is on that account more liable to be 

 lost when the nutrition of the sensory cell is interfered with. In the normal 

 eye the peripheral zone of the retina is totally blind to colour. If you turn the 

 right eye outwards, close the left, and then move a strip of coloured paper from 

 the left to the right in front of the nose, the image of the paper will first fall on 

 the peripheral zone of the retina, and its form will be seen, though indistinctly, 

 but not its colour. It is difficult to say iu that case whether the colour-blindness 

 is due to the state of the retina or to that portion of the vision centre in the brain 

 associated with it. The absence of cones from the peripheral part of the retina 

 has been assigned as the cause, but it is much more probable that the portion of 

 the vision centre associated with the periphery of the retina, being comparatively 

 little used, is less highly developed for form sensation, and not at all for colour 

 sense. It is evident that the production of a sense of white or grey in the absence 

 of all colour sense is not to be explained on the theory that it results from a 

 balanced stimulation of red, green, and violet nerve terminals. 



I need scarcely say that colour-blindness has attracted a large share of atten- 

 tion, not only because of its scientific interest, but still more on account of its 



