12.5 cm. focal length. G is a Thorp's replica of Rowland's gratings, 15,000 lines to the inch, giving a spectrum of the slit H apparently in the centre of the field of view of the optical system formed by the lenses D, E, F, which constitutes a kind of microscope, with the lens K serving as condenser. Behind K is placed a screen L of suitable coloured glass, or a train of prisms for illuminating the apparatus with spectral light.
On rotating the shaft the disc A allows a narrow spectrum of about 10" dispersion and 1" wide to appear momentarily, and immediately afterwards a bright flash of monochromatic light fills the field.
As in my paper on Simultaneous Contrast,[1] a strip of black material is pasted on the plane side of the lens E to serve as a back-ground for the spectrum, which should exactly cover it. The part, therefore, of the retina on which the image of the spectrum falls is not directly stimulated by the second flash, which fills the rest of the field. The results show, however, that it is indirectly affected as in Dr. Shelford Bidwell's experiments.
With a screen of ruby glass the spectrum corresponds with that seen during temporary red blindness, i.e., the red is missing, and the orange and yellow are replaced by green. With a blue screen composed of a film stained with Prussian blue, and a pale cobalt, or, better, a gelatine film stained with aniline blue, the blue and violet disappear, and the green is weakened, but the red is bright. The addition to these of a yellow glass causes the violet of the spectrum to reappear. And the same effect is produced by a green glass. In a word, the results are exactly similar to those described in Section II, paragraph 3, of my paper on Successive Contrast, the difference between the two experiments being that in the new one the retinal area on which the contrast effect, or colour-blindness, is produced, not being exposed to the direct action of the second stimulus, it becomes possible to use a series of rapid flashes instead of a single one, and the effect appears practically continuous.
But the difference between this and Dr. Shelford Bidwell's experiment consists in the use by him of a second stimulus compounded of all the colours, and by me of a second stimulus restricted to one colour. The first step in the discussion must be to ascertain the relation between the results obtained when the second stimulus is white, and when the second stimulus is of the same colour as the first.
In every experiment I have made, whether with one double flash or with a series of such flashes, the result has been the production of a considerable degree of momentary blindness to the colour or colours composing the second flash of the cycle. And this statement is equally true if the two flashes are of different colours. Whatever colour-sensations are common to the two flashes are wholly or partly blotted
- ↑ 'Roy. Soc. Proc.,' vol. 67, p. 227.
