956 On Momentum in the Electric Field. 
this pressure of radiation as it is called, was shown by Maxwell 
to be a consequence of the Hlectromagnetic Theory of Light ; 
it has been detected and measured in some beautiful experi- 
ments made by Lebedew * and by Nichols and Hullf. The 
magnitude of this pressure in the case of light waves is 
very small. 
If the light, instead of being absorbed, were wholly reflected, 
then, when light is incident normally on the surface, the 
incident light communicates to it per unit time an amount of 
momentum in the direction of the light equal per unit of 
surface to H, where H is the energy of the incident light per 
unit volume ; the reflected light carries away per unit time 
an amount of momentum in the opposite direction equal to H. 
Hence the effect on the reflector is the same as if it received 
2E units of momentum per second in the direction of the 
incident light; 7. e.,as if the reflecting surface were acted on 
by a pressure equal to 2H: hence for normal incidence the 
pressure on a perfectly reflecting surface is twice that on a 
pertectly absorbing one. If the absorbing body were moving 
towards the wave with the velocity wu, the amount of momentum 
absorbed in unit time, and therefore the pressure of the 
radiation, would be increased in the proportion of V+w to V. 
Radiation will give rise to pressure when it is refracted, for 
the momentum in the medium being in the direction of pro- 
pagation, when the latter is changed the momentum in the 
medium is changed also, and therefore by the principle of 
the Conservation of Momentum the refracting substance 
must suffer a change in momentum. Thus, suppose a beam 
-of light is travelling along a curved path in a refracting 
medium ; then, if ds is an element of the path, d@ the angle 
between the direction of the beam at the beginning and end 
of the element, the direction of momentum is changed by 60 
in passing along ds. Thus, if M is the momentum per unit 
volume in the beam, an amount of momentum VM60@ at right 
angles to the path of the light must be communicated to the 
medium per unit time: thus the force per unit volume at 
right angles to the light will be VM or E/p, where p is 
the radius of curvature of the beam. 
* Rapports présentés au Congrés International de Physique (2) p. 138. 
+ Proc. American Academy of Arts and Sciences, xxxviii. p. 559. 
a | 
