ON LIGHT. 293 



limit. Thus disembarrassed of the complexity of over- 

 lapping rings of several colours, the phenomenon now is 

 studied to greater advantage, and its explanation on 

 either theory is more readily intelligible. That afforded 

 by the corpuscular theory (supplemented by the New- 

 tonian hypothesis of the fits of easy reflexion and trans- 

 mission) is very simple and obvious. These "fits" or 

 phases, it will be remembered, are supposed periodically 

 recurrent i.e., succeed one another, or rather are re- 

 peated over and over again, in the same order and in- 

 tensity, at equal intervals of time. The same phase then 

 will recur to the corpuscules, at equidistant points of 

 space in their progress through any uniform medium (in 

 which the velocity of light is constant). Where the 

 thickness of the film is nil, or so very minute as to bear 

 no comparison to the distance which separates two of 

 the equidistant points, it is obvious that having passed 

 one surface they will still be in a state to pass through 

 another, and will therefore not be reflected, so that in 

 that case the reflected illumination of the first surface 

 will receive no augmentation from light reflected at the 

 second. The same is true if the thickness of the film be 

 exactly that of two such equidistant points, or its double, 

 triple, &c., for in those cases the corpuscule will arrive at 

 the second surface in the same state and with the same 

 dispositions as to reflexion or transmission as at the first ; 

 and therefore, having penetrated the first, will also pene- 

 trate the second. On the other hand, for thicknesses ol 

 the film exactly intermediate between these, the cor- 

 puscuie on arrival at the second surface will be exactly 



