The Able Diffraction Theory. By J. W. Gordon. 377 



in fig. SI, let us next consider what the result must be ; and for this 

 purpose we will take as the simplest possible case a narrow parallelo- 

 gram. Such an object, if we had a true point to work with, would be 

 depicted thus :— StartiDg from one corner, A, fig. 82, we should 

 carry the point along one edge A B of the parallelogram, so that 

 its trail would trace out a line. Next we should carry the line 



Fig. 83. 



along the longitudinal axis of the parallelogram so that its travel 

 would develop a surface, and when that was done the parallelogram 

 would stand disclosed, the integral of the point. By a similar series 

 of operations we may integrate a surface from the antipoint of fig. 81 ; 

 but here a question of orientation arises which did not come into 

 consideration when we were dealing with the absolutely symmetrical 

 mathematical point. For it is obvious that an infinite number of lines 

 can be traced by carrying the antipoint in different directions. Fig. 83 

 makes this plain without verbal ex- 

 planation. Let it now be assumed 

 that the short edge of the parallelo- 

 gram lies in the long axis of the anti- 

 point. Tracing it, we shall then 

 obtain three short lines, as in fig. 84. 

 Now it is to be observed that the 

 flanking lines here are not pure spectra like the flanking images in 

 the antipoint. They are built up from such spectra, which overlie 

 one another in the resulting lines, so that the blue constituent of one 

 combines with the red constituent of another to produce white illumi- 



