Defective color vision and its inheritance.

We owe the first accurate description of color blindness to the chemist John Dalton, who in 1798 characterized his own condition in the words 'persons in general distinguish six kinds of colour in the solar image; namely, red, orange, yellow, green, blue, and purple.... To me it is quite otherwise:-I see only two, or at most three, distinctions. These I should call yellow and blue; or yellow, blue, and purple. MiIy yellow comprehends the red, orange, yellow, and green of others; and my blue and purple coincide with theirs. That part of the image which others call red, appears to me little more than a shade, or defect of light....". In this superb paper, Dalton set all the main themes for the study of color blindness. He pointed out that his vision had always been this way; that his brother had the same defect; that he knew of nearly 20 such cases, all male; and that in all but perhaps one instance the parents and children of such persons had normal vision.' Dalton thought the trouble to be that his vitreous humor was tinted blue, and so did not pass red light; but Thomas Young said of this: ". . . it is much more simple to suppose the absence or paralysis of those fibers of the retina which are calculated to perceive red."2 Dalton held to his own view, and the argument was settled only by autopsy at his death in 1842, when his eyes were found to be normally transparent to red light.3 Color blindness is still called daltonism throughout Europe; but Young's explanation, that it involves the loss of one of three normal color mechanisms, remained for many years the dominant theory of this condition. This is, however, as Maxwell later made plain, more a mathematical than a physiological theory.4' It implies only that normal human color vision involves the operation of three independent variables, hence is trichromatic, with no indication of what those variables might be. The most obvious expression of this condition is that normal persons must mix three different colors in various proportions to match all the hues they see. Most color-blind persons are dichromats: their vision involves the play of only two variables, and they can match all discriminable hues by mixing two colors. There is also a rare olass of monochromats who can match all colors with one another by adjusting the brightness. Finally there are the anomalous trichromats, discovered by Rayleigh,6 who have three-color vision, but whose mixtures differ in proportion from those of normal trichromats and from one another. Trivariance, to be expressed in the end result, must obtain at all levels of the visual process. The intrusion of a fourth variable at any point would be wasted; the reduction to two variables at any point would make the whole system dichromic. One could in principle account for dichromia through the loss of one of the three color-vision pigments; or the three pigments could be distributed between only two types of cone; or all three classes of cone could discharge their impulses through only two nerve channels; or the normal three mechanisms, distinct to that point, could arouse only two sensations. Add to these the possibility of modified colorvision pigments, and one can appreciate the wide latitude of theory that has existed in this field. Most if not all of these changes have been rung in one or another explanation of color blindness. What is more important, among all the varie-