the Spectra of Stars. 449 



A prism was mounted before the object-glass of a telescope 

 fiirnislied with a micrometer; and to this telescope a smaller 

 one was attached, at such an angle, that one observer could view 

 a star by direct vision, while another simultaneously observed its 

 refracted image. The fii'st observer made the star intersect the 

 wires of the small telescope ; the second brought the micrometer 

 wire to intersect a line of the spectrum. The inclination of the 

 optical axes of the telescopes being known, the deviation of the 

 refracted rays was then readily found. 



The difficulty of measuring the deviation of the refracted light 

 of a star, which arises from the apparent diurnal motion of the 

 heavenly bodies, is completely overcome by this mode of obser- 

 vation, but otherwise it is very inconvenient. Other methods of 

 observation would be to refract the light in a vertical plane, and 

 to observe so near the meridian that the star's variation in alti- 

 tude would be insensible ; or to observe with an equatoreal tele- 

 scope, refracting the light in the plane of the declination circle : 

 but the first of these devices is obviously all but impracticable, 

 and the second depends on difficult instrumental adjustments, 

 while both involve corrections for atmospherical refraction. The 

 method I have now to propose is perfectly free from all these 

 objections. 



A prism is placed on a stage, furnished with proper adjusting 

 screws, immediately behind the horizon-glass of a sextant or 

 reflecting circle, and in the prolongation of the axis of the tele- 

 scope through which the star is to be observed ; the whole being 

 mounted on a suitable stand. The observer can thus see, at 

 once, both the image of the star formed by rays which have been 

 reflected at the mirrors of the sextant or reflecting circle, and 

 the spectrum formed by rays which have passed through the 

 prism, which has been adjusted to its position of minimum devia- 

 tion. The image of the star being then made to coincide with 

 any of the lines in its spectrum, the deviation of the rays is ob- 

 tained directly, by reading off the angle indicated by the sextant 

 or circle. 



I have had an apparatus constructed on this principle, but 

 omng to unfavourable weather, I have as yet been able to make 

 only one imperfect observation. This was on Mars, on the 16th 

 of May. Notwithstanding the proximity of the planet to the 

 moon, and the brightness of that luminary, then twelve days old, 

 the spectrum was more brilliant than I anticipated. The sharp 

 disc of the planet's reflected image, as the tangent-screw was 

 turned, glided on the edge of the spectrum like a bead along a 

 thread ; and contact could be made with the utmost nicety, the 

 brightness of the reflected image being reduced to suit that of the 

 spectrum. 



