6 Prof. A. Kundt on the Indices 



suitably cemented upon the plate of the spectrometer for 

 the observations. An arrangement of screens was placed in 

 front of it which, in the case of simple prisms, left free either the 

 prism or a neighbouring portion of the glass, and in the case 

 of double prisms the one prism and the other in succession. 

 The angles of refraction, and in the case of double prisms the 

 sum of the two angles, were determined in the usual manner 

 by reflexion of cross -wires in a Gauss' eye-piece ; the de- 

 flections by adjusting a fine cross of wires on the image of a 

 slit. In both determinations the plate with the prisms re- 

 mained fixed, and the divided circle with the observing tele- 

 scope was turned. The side faces of the glass plate must be 

 plane ; but, as is easily seen, it is a matter of no consequence if 

 these faces have a slight inclination to each other. 



According to the above formula it is sufficient to adjust the 

 glass plate with the prisms at right angles to the incident 

 rays by eye, although generally this adjustment was controlled 

 by reflexion. Since the surfaces of the prisms are very 

 small — 2-3 millim. broad and about 10 millim. high — the 

 pencil of rays received by the objective was very small ; hence 

 the adjustment of the eyepiece of the observing telescope in 

 the right focal plane is more difficult, and the image of the 

 slit, after passage through the thin prism, is never sharp, but 

 has indistinct edges produced by reflexion. In order to 

 avoid the dangerous errors resulting from insufficient adjust- 

 ment of the observing telescope, observations by way of 

 control were made each time, by the aid of which the tele- 

 scope could be correctly adjusted. Into the details of this 

 process it is not necessary to enter. 



The deflection was always determined for white light — lamp, 

 sun, or electric lamp. Notwithstanding that with some of 

 the metals, as will be seen below, the dispersion is very large, 

 yet with the small angle of the prisms used the deflection for 

 white light could be determined exactly. The value observed 

 applies to the mean rays of the spectrum. Then the deflec- 

 tion was determined for red and for blue light, for which 

 purpose either the sun or the electric light was always 

 employed. 



For observations in the red the rays passed through from 

 one to four red glasses, according to the brightness of the 

 source of light and the transparency of the prism ; for the 

 observations in the blue, through a blue glass, and a vessel 

 with more or less concentrated solution of copper oxide 

 in ammonia. Since it was not always possible to employ the 

 same number of glasses and the same concentration of solu- 

 tion, the mean wave-length of the red and blue light used is 

 not the same in all 'the experiments, so that the observations 



