166 



BELL SYSTEM TECHNICAL JOURNAL 



the boundary, this remains the truth for the components of force 

 parallel to the surface; but the components along the normal, applied 

 respectively from the direction towards the air and the direction to- 

 wards the water, need not be perfectly equal. After the corpuscle 

 has gone through the transition region and reached the depths of the 

 water, it continues in a straight line with a momentum of which the 

 component parallel to the boundary — the "tangential" component, 

 say — is still the same as it was in the air, while the normal component 

 is changed. Denote by pt and pn these two components of the original 

 momentum of the particle through the air, by p the magnitude of their 

 resultant which is the original momentum ; by p/, p„' and p' the corre- 



Fig. 2. 



spending quantities for the final flight of the corpuscle through the 

 water. From Fig. 2 we see : 



sin d = Ptfylpt'' + pn" = pt.lp, sin 0' = p/jp, 



and since pt = p/ : 



sin djsin 0' = p'/p. 



(4) 



(5) 



The corpuscle-theory therefore leads to the statement that the sines 

 of the angles of incidence and refraction stand to one another as the 

 momenta of the corpuscle in the first medium and in the second; and 

 when light is refracted towards the normal, the corpuscles must move 

 with a greater momentum in the second medium than in the first. 



