the Interferences of Light, 3 



to a mode of physical action of the most general kind, and 

 their principles have been trietl most successfully by the se- 

 verest test of mathematical evidence. The earlier supposi- 

 tions then appear to have been very little more than the first 

 legitimate steps in the progress of generalization, and to have 

 been made in strict conformity to the laws of the inductive 

 process, which would always lead us to regard a new and un- 

 explained set of phaenomena as effects siii generis, until they 

 can be fairly reduced to some existing and established genns^ 

 of a more comprehensive character. 



(4.) The phgenomenon of diffraction may be exhibited on a 

 minute scale by candle light, with no other apparatus than a 

 small lens having a piece of fine wire stretched across in con- 

 tact with its suriace. Holding the other surface next the eye, 

 if we look through the lens at the flame of a candle at some di- 

 stance, or, what will be still better, at its light admitted through 

 a narrow slit, the wire being parallel to the flame or line of 

 light; the dark image of the wire will be seen edged by the 

 external fringes, and the shadow marked with the internal 

 stripes in a remarkably beautiful and distinct manner. 



(5.) The fringes of an aperture (which are no other than 

 two sets of external fringes overlapping) are seen with great 

 splendour if we merely place a narrow slit near the flame of 

 a candle, and view it through another held close to the eye, 

 parallel to it, at the distance of a few feet. 



(6.) Beautiful sets of colours (the theory of which is evi- 

 dently dependent on interferences) are seen on viewing a 

 candle, or line of light, by very oblique reflection from any 

 moderately polished plane surface, as ivory, ebony, &c. held 

 close to the eye. 



The images of a narrow line of candle light, formed by the 

 successive internal reflexions of a piece of glass slightly pris- 

 matic, may be seen divided into several dark and bright 

 bands. 



These last experiments, though of so easy and familiar a 

 nature, do not seem to have been generally noticed. 



(7.) The experiment originally proposed by Fresnel, for 

 showing the phitnomenon of interference by reflection from 

 two plane glasses inclined at a very large angle, has been re- 

 garded as one of great delicacy to exhibit, and requiring a 

 very nicely adjusted apparatus. This idea, however, seems 

 to me unfounded ; for if care be taken to provide two halves 

 of the same piece of glass about one inch square, nothing is 

 easier than to produce the efl'cct by merely laying them flat 

 on a smooth table with the cut edges in contact (or even 

 nearly so), and giving one of them a slight inclination by 

 B 2 placing 



