CLASSICAL THEORY OF LIGHT 751 



a beam of light from a flame or an arc consists of myriads of feeble 

 beams each proceeding from a single atom. Each is divided — the 

 methods of division are the methods of producing interference-fringes — - 

 and the separate parts are then caused to overlap. Each pair which 

 came originally from a single atom produces a set of interference- 

 fringes, and the fringes for all these pairs coincide in space. Each 

 fraction of a divided beam may also interfere with a fraction f)f 

 another, proceeding from another atom; but owing to the uncontrolled 

 and uncontrollable phase-difi^erences between the beams of a pair so 

 formed, the fringes for these pairs do not coincide, and on the whole 

 they efface one another. By the quantum-theory the explanation — 

 not indeed of the fact that interference occurs only under these special 

 conditions, but of the fact that it ever occurs at all — is not so easy. 

 Indeed the fact commonly expressed by saying that light from a 

 source is "coherent" with itself, has been regarded as the most difficult 

 of all for the quantum-theory to explain. 



To produce interference, then, we must divide a beam of light and 

 cause its parts to cross each other's paths. The simplest of the devices 

 which effect this were invented by Fresnel; a pair of prisms which 

 turn two portions of the beam towards one another, and a pair of 

 mirrors which reflect two portions across each other's routes. A single 

 mirror indeed suffices; standing acoustic waves are produced thus, 

 in Kundt's tube and otherwise, with values of the angle 26 sometimes 

 as great as 180°; but light-waves are so short that with so great an 

 angle the distance between dark fringes would be too small to measure, 

 if not indeed to perceive ; and we must use the facility for expanding 

 them which the factor sin 6 in equation (15) offers us. 



The Interferometer 

 In the devices which I have thus far mentioned, the interference 

 of overlapping wave-trains oblique to one another causes the formation 

 of alternate zones of darkness and light in space; and the visible fringes 

 are the cross-sections of these zones upon a screen set up to intersect 

 them. There are however other instruments in which the overlapping 

 beams are parallel to one another, the region which they occupy is 

 not traversed by bands of light and shade, and a screen thrust across 

 it shows uniform illumination ; and yet when the eye or the camera is 

 located in that region, fringes are produced upon the retina or on the 

 plate by the action of the lens of either. These are not so easily under- 

 stood as the earlier devices, and yet it is important to comprehend 

 them, for the striking applications of interference have been made by 

 means of such as these. Among them are the interferometer of Fabry 

 and Perot, and that of Michelson. 

 48 



