14 Stokes, on the Nature of the Rontgen Rays. 



will spread out spherically from the place of disturbance.* 

 You might at first sight suppose that you could have 

 a wave, in any limited region of which you might have 

 a transversal disturbance in some one direction, the same 

 all through the thickness of the shell occupied by the 

 wave, though naturally the direction of disturbance might 

 vary from one region to another more or less distant 

 region. But the dynamical theory shows that that is 

 not possible. In any limited region, or elementary area, 

 as we may regard it, of the wave, as you pass in a 

 direction perpendicular to the front, the disturbance in 

 one direction must be exchanged for a disturbance in the 

 opposite direction, in such a manner that ultimately — 

 that is, when the radius of the wave is very large com- 

 pared with its thickness — the integral of the disturbance 

 in one direction, which we may designate as positive, 

 must be balanced by the integral of the disturbance in 

 the opposite, or negative, direction. The simplest sort 

 of " pulse," as I will call it, in order to distinguish it from 

 a periodic undulation, would be one consisting of two 

 halves in which the disturbances were in opposite direc- 

 tions. The positive and negative parts are not necessarily 

 alike, as one may make up by a greater width, measured 

 in the direction of propagation, for a smaller amplitude ; 

 but it will be simplest to think of them as alike, except 

 as to sign. The following figure represents this concep- 

 tion, the positive and negative halves being distinguished 

 by a difference of shading. 



r<<<<<«€<<<<<^ 



* If the medium be compressible there will be two waves, that which 

 travels the more swiftly consisting of normal vibrations ; but the opinion 

 has already been expressed that it is transversal vibrations with which 

 we are concerned. 



