466 HISTORY OF SCIENCE. 



is in fact the position of the darkest part of a dark band. Such being 

 the theory, it is perfectly easy for a geometrician to calculate the 

 intervals which ought to be found between the bands at any given dis- 

 tances from the wire. It is indeed obvious by simple inspection of Fig. 

 208 that the position of any of the intersections of the circles is found 

 when we know the radii of the circles, as their centres s, A, and B are 

 fixed points. Now, for calculating the positions of the bands according 

 to this theory, it is only necessary to measure the distances between 

 the fixed points and to know the wave-length of the rays of light em- 

 ployed. The radii of the circles will be always multiples of the half- 

 wave-length. Adopting the measures deduced by Newton from the 

 phenomena of his rings (page 224) as representing the wave-length, 

 Fresnel calculated what ought to be the distance of the dark bands 

 within and without the geometrical shadow of the wire at certain posi- 

 tions where he had measured the actual distances between the bands. 

 The close agreement between the calculated and the observed dis- 

 tances will be best shown by the tables below from Fresnel's paper. 

 The light used was homogeneous, that is, of one colour, and was ob- 

 tained by permitting sun-light to first pass through a particular kind 

 of red glass. Had white light been admitted, the effects of interfe- 

 rence would have manifested themselves not in the changes of mere 

 intensity, which give rise to the dark and light bands, but in certain 

 phenomena of coloration. It is obvious that only red rays can in- 

 terfere with red rays, and that, for instance, green rays could never 

 interfere with red, but only with green. The colours which arise from 

 interference when white light is used, are due to the extinction of the 

 complementary colour. Thus, for example, white light gives on a screen 

 a green band in the place where red light would produce a dark band. 

 The wave-length corresponding to the light transmitted by the red 

 glass, deduced from Newton's results, was '000623 millimetres. As 



Indies. 



the distances in Fresnel's table are given in terms of the metric system. 

 we shall here insert a scale of the actual dimensions of a decimetre or 

 one-tenth of a metre, with its division into centimetres and millimetres, 

 Fig. 209. 



