206 FRESNEL. 



is not every inequality of this kind which will necessa- 

 rily produce a destruction of light ; such difference may, 

 on the contrary cause the rays to reinfore each other. 



But when we know what is the hast difference of 

 route gone through, at which the rays may be super- 

 posed ivithout influencing each other, we then obtain all 

 the other differences of route which give the same result 

 in a very simple manner ; for it suffices to take the 

 double, the triple, the quadruple, &c., i. e. every whole 

 multiple, of the first number to give them. 



If we have noted in like manner the least difference 

 of route which produces complete destruction of the two 

 rays, every odd whole multiple of this first number will 

 also be the indication of a like destruction. 



As for differences of route which are not numerically 

 comprised either in the first or in the second of the 

 above series, they correspond only to partial destruc- 

 tions of the light, or mere weakening of its intensity. 



These series of numbers, by aid of which we can tell 

 whether two rays at the moment of intersection ought to 

 interfere or merely to combine without influencing each 

 other, have not the same values for the differently col- 

 oured rays ; the smallest values belong to the violet rays, 

 the greatest to the red, and the intermediate values to 

 the intermediate rays. It results, that if two white rays 

 cross at a certain point, it may be possible that in the 

 infinite series of differently coloured rays of which that 

 light is composed, the red, for example, alone may be 

 destroyed and disappear, and thus the point of concourse 

 may appear green, as being the white light deprived of 

 its red component. 



Interference, then, which in homogeneous light pro- 

 duces only changes in intensity, will manifest itself when 



