242 ON ACCIDENTAL OR SUBJECTIVE COLORS. 



rapid and alternating movement, perpendicularly to tlie direction of its clefts, 

 the broken parallels will be perceived with particular distinctness. 



Phenomena analogous to those just described may be seen by observing 

 black and parallel lines traced on white paper, like those which represent the 

 sea on an engraved chart, or by looking at the sky through the luminous inter- 

 vals left between parallel threads. If the eye observes these objects fixedly 

 and continuously, the black lines soon lose their straight direction and their 

 parallelism, and present luminous spaces circumscribed nearly like rings of a 

 certain number of links. When this change takes place, the eye which ob- 

 serves it feels a sensation of uneasiness — an effect which is produced also in the 

 closed eye. When this dazzling effect supervenes the luminous spaces between 

 the broken lines become colored, some with yellow, others with blue and green. 



The phenomena produced in these two experiments pertain naturally to rec- 

 tilinear undulations propagated on the retina, and the interference and crossing 

 of the undulations, in consequence of which the black lines are broken into de- 

 tached portions and the colors are produced, spring from the want of fixedness 

 in the head and hand, which causes a parallelism in the successive undulations. 



The action on the retina of small and dilating luminous points produces very 

 interesting phenomena. If we observe the sun through a small opening situ- 

 ated at a great distance from the eye, or the small image of the sun formed by 

 a convex lens or a concave mirror, or seen on a convex surface, the light which 

 falls on the retina does not form a clear and definite image of the luminous 

 point, but emits in all directions an infinity of rays which cover, in certain cases, 

 almost the whole retina. These rays are extremely brilliant and sometimes ac- 

 companied by colors of singular variety and beauty. The glittering point 

 propagates around itself circular undulations which are broken and colored by 

 interference, and which, being in constant movement from the centre of the re- 

 tina in all directions, occasion the irradiations above indicated. This curious 

 phenomenon appears in all its splendor when we look at the luminous centre 

 which results from the ignition of a jet of oxygen and hydrogen gas on a piece 

 of chalk, or the combustion of points of charcoal at the poles of a strong bat- 

 tery. HoAV often have we contemplated with surprise this circular zone of 

 sparkling rays which appear under the form of discontinuous right lines scat- 

 tered at equal distances from points by turns excessively brilliant and obscure. 



If we observe through a narrow opening the luminous centres just described, 

 a very singular effect is produced. A vortex of circular rays exhibits itself on 

 each side of the brilliant point, having a rapid rotary movement. This remark- 

 able configuration of the rays is evidently produced by the union of a system of 

 parallel undulations with a system of spherical or circular waves ; the intersec- 

 tions of the parallel fringes and divergent irradiations form the circular rays as 

 in the case of ordinary caustics. All these phenomena, adds Sir D. Brewster, 

 •whatever be their true cause, evidently show that the light which falls on the 

 retina exerts an action even on the parts which do not receive it directly, and ^ 

 that the same action renders other parts of the retina insensible to the light 

 which actually falls upon them. 



M. Peclet also has made an observation not less curious. If the sky be ob- 

 served through a fissure half a millimetre in width dark stripes will be perceived 

 irregularly distributed in the opening, but always in the same manner, whatever 

 may be the size, form, and nature of the illuminating body ; they do not change 

 even when this body is sufiiciently slender to yield fringes by diffraction ; and 

 Avhen the opening is covered by a colored glass, the stripes seen through the 

 glass are on the prolongation of those which are formed in the free part of the 

 fissure. These stripes change place and become weaker when the width of the 

 opening is increased ; beyond a millimetre only very indistinct ones are per- 

 ceptible near the borders. If the fissure is more remote, the stripes become less 

 numerous, and entirely disappear at the distance of distinct vision. When a 



