310 ON LIGHT. 



another of expansion by which they return to their places, 

 these waves, 256 in number, being all comprised in, and 

 exactly rilling the distance (1090 feet) run over by sound 

 in that time, each entire wave will occupy 4-254 ft. 

 And, vice versa, if we knew this a priori to be the wave- 

 length, we should rightly conclude 256 to be the num- 

 1 ber of complete vibrations or pulses per second. 



(94.) Mechanical processes enable us to grind and 

 polish a glass surface into the segment of a sphere of 

 any required radius as well as to a plane almost mathe- 

 matically true. Suppose such a glass surface worked, 

 we will say, to a sphere of 100 feet radius, to be laid 

 (convexity downwards) on a truly plane glass. The 

 coloured rings will be formed, as above described, about 

 a central dark spot; and if illuminated, instead of 

 ordinary daylight, by the prismatic rays, in succession, 

 a series of simply bright and dark rings of the several 

 colours in their order will be formed, whose diameters in 

 different series will correspond to their respective tints. 

 Under these circumstances, the linear measurement of 

 these diameters may be performed with ease and with 

 great precision. Now these diameters are the chords of 

 arcs of a circle on a radius of 100 feet represented in fig. 

 7, by the horizontal lines, the versed sines of whose halves 

 corresponding (represented by the perpendicular lines) are 

 the distances between the glasses at those points, or the 

 thicknesses of the interposed film of air, and are easily 

 calculated when the radius and the chords are known. 

 On executing the measurements it is found that these 

 distances, reckoning outwards and commencing with the 



