REPORT ON REFRACTIONS OF SOLAR SPECTRUM. 3 



papers^ taking the mean where the different sets were compara- 

 ble ; and in other cases adopting those which appeared to me 

 most to be relied on. 



My observations were made with an apparatus, the essential 

 parts of which are a graduated circle, having the prism at its 

 centre, and a small achromatic telescope with cross wires, and 

 a power of 10 nearlj^, directed to the prism, and moveable on 

 an arm about the centre, along with the index. The diameter is 

 10 inches ; the limb is divided on silver to 10', and by two oppo- 

 site verniers with lenses to 10". It was originally made and di- 

 vided by Allan, but fitted up for this special purpose, under my 

 directions, by Mr. Sirnms. The whole will be directly under- 

 stood by inspection of the annexed plate. (See Plate I.) 



The slit, which is the origin of light, is about g^jth of an inch 

 broad, formed by the edges of two brass plates made by Mr. 

 Simms, and inserted in a screen, outside of which is the usual 

 apparatus for throwing the sun's rays into any convenient di- 

 rection ; the prism is placed at about 12 feet distance. 



The absolute deviation of each ray is thus observed directly 

 from the zero point, or that which is shown on directing the te- 

 lescope to the slit. The adjustment for parallelism in the edge 

 of the prism with the slit is easily made, and the position of 

 least deviation found accurately by the cross wires and fixed 

 lines. 



From the observed deviations, the indices are deduced by the 

 well-known formula {B being the minimum deviation, t the 

 prism-angle, and fi the index,) 



log. fj, = log. sin. ~ log. sin. y. 



For liquids, hollow prisms or troughs of truly parallel plate-glass 

 were employed, the angle of each being determined by a me- 

 thod described in my second paper; a thermometer was inserted, 

 and the temperature of the medium always noted. 



In the course of my observations some doubt had arisen as to 

 the exact identi/icafion of certain of the standard rays, according 

 to Fraunhofer's designation of them, owing to the very defective 

 representations given of them in various optical treatises, which 

 fail to convey the peculiar characteristics which mark the differ- 

 ent bands. 



Among the larger maps of the spectrum, that in the Edin- 

 burgh Encyclopaedia (art. Optics, plate 433, fig. 16.) is profess- 

 edly copied from Fraunhofer's, which is given in Schumacher's 

 Journal, before referred to, (tab.ii. fig. 6.) ; and this, taken from 

 that in the Munich Transactions (tab. ii. fig. 5.). This last 

 b2 



