as an exjperimentum crucis as to the nature of LigJit. 38 1 



light; and it must be confessed that the latter theory was 

 greatly advanced in probability by the demonstration of the 

 former. Certain circumstances, however, such as the colour 

 and arrangement of the bands, are required to be examined 

 before we can consider the primary theory to be confirmed in 

 the same extent as its subordinate theory. For instance, ac- 

 cording to the fundamental property of wave interferences 

 the central band must be white; but if it should be found in 

 the experiment that the central band is black, this discre- 

 pancy, whilst it would not weaken the demonstration of the 

 theory of interference, would yet be a fatal objection to the 

 theory of undulations. 



Now if two equal series of circular or spherical waves 

 which have the same direction, or nearly so, arrive at any 

 points in the transmitting fluid, in such a manner that the 

 like parts of the waves arrive at the same instant, then their 

 conjoint effect will be to produce a resultant wave stronger 

 than either of the component waves. If, however, the two 

 series are in a state of complete discordance, so that one series 

 would produce an effect equal and opposite to that which would 

 be produced by the other series, they would counteract each 

 other, that is, no resultant wave would be produced. Two 

 series of waves, which are respectively of equal diameters, and 

 similar, at every instant may be denominated simultaneous ; 

 and if we draw a line bisecting perpendicularly that joining 

 their origins, the waves of each series will meet in the same 

 state everywhere along this line, since every point in it is 

 equally distant from the two origins. When these waves are 

 very distant from the origins compared with the distance of 

 the origins from each other, they will have very nearly the 

 same directions, and we shall have resultant waves of greater 

 strength than the component waves. This is a conclusion de- 

 pending only on the fundamental properties of waves, and 

 does not involve any review of the various hypotheses which 

 are held, as to the nature of the vibrations of the particles of 

 the fluid through which the waves are propagated. 



The experiment with the two mirrors, before mentioned, 

 produces the case we have just discussed ; for the single pencil 

 diverging from the original luminous point, is made into two 

 pencils after reflection by the mirrors; and these two, if light 

 consist of waves, are composed of waves both equal and si- 

 multaneous. We have here then an experiment on which to 

 test the wave theory of light, and the experiment itself shows 

 us which is the band corresponding to the line bisecting per- 

 pendicularly that which joins the luminous images of the 

 point. For if the luminous point be formed of white light, 



