8 M. Fraunhofer on Coloured Spectra from Flame, 



instruments entirely for this purpose. With a telescope hav- 

 ing an object-glass four inches diameter, I have obtained from 

 it several interesting results, although the experiments are by 

 no means finished. The flint-glass prism of this instrument 

 has an angle of 37° 40', and the same diameter as the object- 

 glass. The angle which the ray falling on the prism forms 

 with that coming out of it is about 26°, so that if the refran- 

 gibility of light of one star were ever so little different from 

 that of another the difference could still be very easily obser- 

 ved. In order to observe and determine with accuracy this 

 difference, if it does exist, I have applied another smaller tele- 

 scope, which is fastened to the other, and crosses it at an angle 

 of about 26°, namely, under the angle formed by the ray fall- 

 ing on the prism, and the ray proceeding from it ; the entrance 

 of the star is observed at the wire of the smaller telescope 

 without a prism by one observer, whilst another notices the 

 entrance of a part of the spectrum of the same star through 

 the larger telescope, which is furnished with a micrometer 

 screw, the moveable edge of which the observer, by means of 

 the screw, places in such a manner that at the moment when 

 the star passes across the wire of the smaller telescope without 

 a prism, one of the fixed lines of the spectrum shall intersect 

 the edge in the larger telescope. 



The instrument, without altering the micrometer, is then 

 directed to another star, to enable us to discover whether its 

 light has the same refraction. If at the moment when this 

 star intersects the wire of the lesser telescope the same colour 

 of the spectrum, or the same fixed line, is at the edge of the 

 micrometer of the larger telescope, the refraction of these two 

 sorts of light is equal. As these experiments require two ob- 

 servers, Mr Soldner, the astronomer, had the goodness to 

 make them with me. These experiments, however, are, as I 

 have said, only to be considered as commenced, and I must 

 yet make essential alterations in the instrument, in order to ob- 

 tain still greater accuracy, and also to gain more time for ob- 

 servation. 



We have not hitherto found ajixed star, the light of which, 

 in regard to refraction, differs perceptibly from the light of 

 the planets. Whenever the fixed lines of the spectra are seen 



4 



