produced ly Iwo Plates of Glass of equal Thickness. 357 



depends upon the action of all the four surfaces of the two 

 plates of parallel glass. 



All the preceding experinieuts were made vvith plates which 

 were cut out of the same piece of glass, and had therefore the 

 same thickness. I now tried plates of different thicknesses, 

 both when ground parallel, and when cut from common plate 

 glass ; but I could never render the coloured fringes visible, un- 

 less when the glass was parallel, and exactly of the same thick- 

 ness in both plates. I also tried plates of topaz, of equal thick- 

 nesses, and plates of sulphate of lime; but though I used pieces 

 of various thicknesses, I have riever succeeded in making them 

 exhibit the coloured fringes, owing, perhaps, to the imperfect 

 flatness of their surfaces. 



In order to ascertain if the magnitude of the fringes depended 

 on the thickness of the glass plates, I procured a piece of parallel 

 crown glass yWo'^'^'^s of an inch thick, and compared the fringes 

 which it produced, at an inclination of 2^ 20', and at a vertical 

 incidence, with those produced by another piece of glass y'^\'^dths 

 of an inch thick. In the first case, the circular image was 

 crossed by five fringes, and in the second case with seven 



fringes: but 121 1(58 ^ » , 



; : : 5 : 7 nearly, 



1000 1000 



In another experiment, I found, from a mean of five measure- 

 ments, that the thickest of these pairs of plates produced fringes 

 each of which had a breadth of 1 1' 10", when the inclination of 

 t!ie plates was 1° 5S'. Now the other pair of ])lates gave fringes 

 13' 3" broad, at an inclination of 2" 20", which gives 13' 29" 

 for their breadth at an angle of I" 5S', and 



1^: A-*^- :: 11' 10": 15' 30". 

 1000 1000 

 Hence the ma^niludes of the fringes are inversely as the thick- 

 nesses of the plates ivhich produce them, at a given incVmation ; 

 and in general the magnitudes of the fringes are in the com- 

 pound inverse ratio of the thickness of the plates, and of their 

 angle of incUnafion. 



Ilitherto we have supposed the glass plates to be placed be- 

 tween the eye and the luminous object, so that only the 2d, 4th, 

 and 6th redected images were seen. When the eye is placed 

 between the plates and the luminous object, so as to perceive 

 tiie 1st, 3d, and 5th reflected images, the coloured fringes are 

 also seen, having the same characters as those already noticed. 



The phsMiomena which have been described are e({ually pro- 

 <lnced when the fringes are formed by polarized light, and they 

 do not suffer the least change when examined by doubly re- 

 fracting or doubly polarizing crystals. 



Z 3 Whoa 



